The rapid advancements in face forgery techniques necessitate that detectors continuously adapt to new forgery methods, thus situating face forgery detection within a continual learning paradigm. However, when detectors learn new forgery types, their performance on previous types often degrades rapidly, a phenomenon known as catastrophic forgetting. Kolmogorov-Arnold Networks (KANs) utilize locally plastic splines as their activation functions, enabling them to learn new tasks by modifying only local regions of the functions while leaving other areas unaffected. Therefore, they are naturally suitable for addressing catastrophic forgetting. However, KANs have two significant limitations: 1) the splines are ineffective for modeling high-dimensional images, while alternative activation functions that are suitable for images lack the essential property of locality; 2) in continual learning, when features from different domains overlap, the mapping of different domains to distinct curve regions always collapses due to repeated modifications of the same regions. In this paper, we propose a KAN-based Continual Face Forgery Detection (KAN-CFD) framework, which includes a Domain-Group KAN Detector (DG-KD) and a data-free replay Feature Separation strategy via KAN Drift Compensation Projection (FS-KDCP). DG-KD enables KANs to fit high-dimensional image inputs while preserving locality and local plasticity. FS-KDCP avoids the overlap of the KAN input spaces without using data from prior tasks. Experimental results demonstrate that the proposed method achieves superior performance while notably reducing forgetting.

Recently, Multimodal Large Language Models (MLLMs) have demonstrated their immense potential in computer-aided diagnosis and decision-making. In the context of robotic-assisted surgery, MLLMs can serve as effective tools for surgical training and guidance. However, there is still a deficiency of MLLMs specialized for surgical scene understanding in endoscopic procedures. To this end, we present EndoChat, an MLLM tailored to address various dialogue paradigms and subtasks in understanding endoscopic procedures. To train our EndoChat, we construct the Surg-396K dataset through a novel pipeline that systematically extracts surgical information and generates structured annotations based on large-scale endoscopic surgery datasets. Furthermore, we introduce a multi-scale visual token interaction mechanism and a visual contrast-based reasoning mechanism to enhance the model's representation learning and reasoning capabilities. Our model achieves state-of-the-art performance across five dialogue paradigms and seven surgical scene understanding tasks. Additionally, we conduct evaluations with professional surgeons, who provide positive feedback on the majority of conversation cases generated by EndoChat. Overall, these results demonstrate that EndoChat has the potential to advance training and automation in robotic-assisted surgery. Our dataset and model are publicly available at https://github.com/gkw0010/EndoChat.

The overfitting of domain signals results in poor domain generalization of face anti-spoofing. The current methods usually improve the diversity of source domains to alleviate this overfitting. However, this benefit is minimal, as even the most diverse domain signals will also be absent in the target domain. In this work, we propose a Domain-Guided Prompt Distribution Learning (DGPDL) built on Vision-Language Models like CLIP, which explores a unified representation of domain signals as a prompt across the source and target domain to alleviate the understanding bias caused by domain gaps. Specifically, we first define a learnable Domain-Specific Distribution (DSD) that covers as many domain elements as possible, such as image quality, color tone, camera settings, etc., which establish connections between different domains and linearly combinable prompt in any domain; Then, based on the style statistics of the given sample, we construct its optimal Domain-Specific Prompts (DSPs) from the defined DSD through the designed Prompt Assemble Attention (PAA) with the similarity matching; Finally, the assembled DSPs will act as carrier or agent to perform on both the vision and language branches, synergistically improving the model's recognition of domain signals. By using the prompt to represent domain signals uniformly, if the model can be robust to DSPs in the source domain, it should be applicable to target domain, as they share the same DSD. By representing domain signals as prompts rather than instantiation features, DGPDL effectively reduces the reliance on specific domain appearances. This design enables the model to dynamically adapt to unseen target domains without the need for retraining. Extensive experiments show that the DGPDL is effective and outperforms the state-of-the-art methods on several cross-domain benchmarks.

Attributing to the strong reasoning capability, behavior models in artificial intelligence for games play a crucial role in creating gaming experiences. For further enhancing the reasoning capability of behavior models, we propose a novel Switch, Reason, and Revise (SRR) framework, which integrates them with Large Language Model (LLM). The SRR framework contains three core components. The component of Dual-Track Experiential Reasoning fully utilizes the agent experiences for reasoning. The component of Block-Retrieval-Augmented Thoughts adaptively determines the granularity of information retrieval for external sources. The component of Self-Reliant Thinking System Switch increases the reasoning speed by model switching and performs the model switching automatically upon the LLM. Together, these three components can strengthen the agent reasoning capability in complex tasks. Experimental results in the Pokémon battle environment demonstrate the effectiveness and efficiency superiority of SRR over the rival methods. Furthermore, we conduct an exploratory study to reveal the potential of the SRR-empowered agent for guiding new players in Pokémon battle games.

Current adversarial attacks pose a serious threat to the robustness of visual-language models (VLMs), including vision-language pre-trained models (VLPMs) and multimodal large language models (MLLMs). Traditional adversarial attacks are example-specific and rely on specific datasets. This practice suffers from low transferability and additional computation cost, while universal adversarial perturbations (UAPs) offer example-agnostic solutions by generalizing across inputs. However, current UAP methods mainly target VLPMs, demonstrating limited transferability and effectiveness in MLLMs. To bridge this gap, we propose the Recursive Perturbation Attack (RPA), a novel black-box UAP method for both VLPMs and MLLMs. RPA employs a recursive perturbations strategy, utilizing token filtering and polynomial sampling methods to generate perturbations, thereby achieving incremental disruption and enhancing the transferability of the attack. To further enhance the effectiveness of the attack, RPA integrates a three-tier modality decoupling strategy, disentangling intra-modal, cross-modal, and fusion-modal features to effectively disrupt feature alignment and interactions. Extensive experiments validate that RPA achieves superior attack performance compared to existing UAP approaches. This work highlights new security concerns in multimodal AI systems and provides insights into the design of more robust models. Code is available at https://github.com/chilljudaoren/RPAttack.

In the present environment where privacy protection is increasingly emphasized, source-free unsupervised domain adaptation (SFUDA) has garnered more attention compared to standard unsupervised domain adaptation (UDA). It concentrates on transferring knowledge directly from well-trained source models to unlabeled target domains without requiring the involvement of source domain like UDA, greatly enhancing data protection capabilities. Many existing methods employ pseudo-labeling to guide this process, but due to domain shift, pseudo-labels often introduce significant noise. Although there are methods to filter out this noise and mitigate its impact, they may also result in the loss of crucial sample knowledge, leading to performance deterioration. In contrast, we propose a novel approach called Progressive Curriculum Learning with Teacher-Student Collaboration (PCTSC) method to mitigate the adverse influence of noisy labels in SFUDA. Inspired by curriculum learning, PCTSC assesses samples’ learning difficulty and trains models in an incremental manner from easy to hard, thereby enhancing the capability of model to against noise. Furthermore, PCTSC employs a two-stage learning approach: initially, a teacher model directs the student model, and later, the student model transitions to independent learning. We assess the effectiveness of PCTSC by conducting extensive experiments across three benchmark datasets, demonstrating its robustness against pseudo-label noise in SFUDA setting.

Unsupervised domain adaptation aims to transfer knowledge from a labeled source domain to an unlabeled target domain. Existing methods, whether based on distribution matching or self-supervised learning, often focus solely on classifying individual source samples, potentially overlooking discriminative information. To address this limitation, we propose FlowUDA, a novel plugin method that enhances existing UDA frameworks by constructing semantically invariant flows from individual source samples to corresponding target samples, forming cross-domain trajectories. By leveraging a diffusion network guided by ordinary differential equations, FlowUDA ensures these flows preserve the topological structure of the source domain, maintaining their distinguishability. Our method then classifies these flows by sampling points along them and transferring labels from source samples, effectively capturing spatial relationships between domains. In essence, FlowUDA transforms the traditional point-based classification on individual source samples into flow-based classification on flows, allowing the model to learn richer, more discriminative features that bridge the gap between source and target domains. Extensive experiments on standard benchmarks demonstrate that integrating FlowUDA into existing UDA methods leads to notable performance gains, highlighting its effectiveness in addressing domain shift challenges.

With the rapid development of Large Language Models (LLMs), an increasing number of Large Visual-Language Models (LVLMs) have achieved unprecedented performance in response generation. Recent work shows that LVLMs are vulnerable to adversarial attacks. However, many existing methods tend to overfit to the source model by overemphasizing specific features, which compromises their transferability. Other approaches suffer from reduced attack effectiveness due to insufficient differentiation between features. In this paper, we propose a novel transfer-based black-box untargeted attack—Shared Adversarial Feature (SAF) dynamic attack. By exploring the feature extraction patterns of LVLMs, we identify the features shared among various models that are most susceptible to adversarial attacks and disrupt them. Moreover, due to the powerful attention mechanisms of LVLMs, they are still able to extract similar semantics from perturbed images, even when primary features are disrupted. We design a dynamic update strategy to address this challenge. Finally, from the perspective of SAF, we conduct an in-depth analysis of vulnerabilities in the vision encoder and projector within LVLMs and find that attacking the projector exhibits stronger transferability across heterogeneous model architectures. Extensive experiments show that our method exhibits superior attack performance compared to existing methods across different models, datasets, and tasks. The code will be publicly available after publication.

3D Gaussian Splatting has emerged as an alternative 3D representation for novel view synthesis, benefiting from its high-quality rendering results and real-time rendering speed. However, the 3D Gaussians learned by 3D-GS have ambiguous structures without any geometry constraints. This inherent issue in 3D-GS leads to a rough boundary when segmenting individual objects. To remedy these problems, we propose SAGD, a conceptually simple yet effective boundary-enhanced segmentation pipeline for 3D-GS to improve segmentation accuracy while preserving segmentation speed. Specifically, we introduce a Gaussian Decomposition scheme, which ingeniously utilizes the special structure of 3D Gaussians, finds out, and then decomposes the boundary Gaussians. Moreover, to achieve fast interactive 3D segmentation, we introduce a novel training-free pipeline by lifting a 2D foundation model to 3D-GS. Extensive experiments demonstrate that our approach achieves high-quality 3D segmentation without rough boundary issues, which can be easily applied to other scene editing tasks. Our code is publicly available at https://github.com/XuHu0529/SAGS.

Remote sensing image fusion aims to create a high-resolution multi/hyper-spectral image from a high-resolution image with limited spectral information and a low-resolution image with abundant spectral data. Recently, deep learning (DL) techniques have shown significant effectiveness in this area. Most DL-based methods approach image fusion as a 2D problem by encoding spectral information into feature map channels. However, our research suggests that this strategy introduces notable spectral distortions. In contrast, some methods consider spectral data as an additional dimension, utilizing standard 3D convolutions to preserve spectral information. Nevertheless, in a standard 3D convolutional layer, the same set of kernels is applied across all input regions, which we have found to be sub-optimal for image fusion. Furthermore, standard 3D convolutions necessitate substantial computational resources. To address these challenges, we propose a novel convolutional paradigm called Adaptive 3D Convolution (Ada3D) for remote sensing image fusion. Ada3D applies a unique set of 3D kernels to each input voxel, enabling the capture of fine-grained details. These adaptive kernels are generated through a two-step process: 1) spatial and spectral kernels are derived from their respective image sources and 2) these two types of kernels are then combined to form content-aware 3D kernels that effectively integrate spatial and spectral information. Additionally, adaptive biases are introduced to enhance the convolutional outcome at the voxel level. Furthermore, we incorporate the group convolution technique to reduce computational complexity. As a result, Ada3D offers full adaptivity in an efficient manner. Evaluation results across five datasets demonstrate that our method achieves state-of-the-art (SOTA) performance, underscoring the superiority of Ada3D. The code is available at https://github.com/PSRben/Ada3D.

Vision-Language Pretrained (VLP) models exhibit strong multimodal understanding and reasoning capabilities, finding wide application in tasks such as image-text retrieval and visual grounding. However, they remain highly vulnerable to adversarial attacks, posing serious reliability concerns in safety-critical scenarios. We observe that existing adversarial examples optimization methods typically rely on individual features from the other modality as guidance, causing the crafted adversarial examples to overfit that modality's learning preferences and thus limiting their transferability. In order to further enhance the transferability of adversarial examples, we propose a novel adversarial attack framework, I&CA (Individual & Common feature Attack), which simultaneously considers individual features within each modality and common features cross-modal interactions. Concretely, I&CA first drives divergence among individual features within each modality to disrupt single-modality learning, and then suppresses the expression of common features during cross-modal interactions, thereby undermining the robustness of the fusion mechanism. In addition, to prevent adversarial perturbations from overfitting to the learning bias of the other modality, which may distort the representation of common features, we simultaneously introduce augmentation strategies to both modalities. Across various experimental settings and widely recognized multimodal benchmarks, the I&CA framework achieves an average transferability improvement of 6.15% over the state-of-the-art DRA method, delivering significant performance gains in both cross-model and cross-task attack scenarios.

Recent advances in surgical robotics and computer vision have greatly improved intelligent systems' autonomy and perception in the operating room (OR), especially in endoscopic and minimally invasive surgeries. However, for open surgery, which is still the predominant form of surgical intervention worldwide, there has been relatively limited exploration due to its inherent complexity and the lack of large-scale, diverse datasets. To close this gap, we present OpenSurgery, by far the largest video-text pretraining and evaluation dataset for open surgery understanding. OpenSurgery consists of two subsets: OpenSurgery-Pretrain and OpenSurgery-EVAL. OpenSurgery-Pretrain consists of 843 publicly available open surgery videos for pretraining, spanning 102 hours and encompassing over 20 distinct surgical types. OpenSurgery-EVAL is a benchmark dataset for evaluating model performance in open surgery understanding, comprising 280 training and 120 test videos, totaling 49 hours. Each video in OpenSurgery is meticulously annotated by expert surgeons at three hierarchical levels of video, operation, and frame to ensure both high quality and strong clinical applicability. Next, we propose the Hierarchical Surgical Knowledge Pretraining (HierSKP) framework to facilitate large-scale multimodal representation learning for open surgery understanding. HierSKP leverages a granularity-aware contrastive learning strategy and enhances procedural comprehension by constructing hard negative samples and incorporating a Dynamic Time Warping (DTW)-based loss to capture fine-grained temporal alignment of visual semantics. Extensive experiments show that HierSKP achieves state-of-the-art performance on OpenSurgegy-EVAL across multiple tasks, including operation recognition, temporal action localization, and zero-shot cross-modal retrieval. This demonstrates its strong generalizability for further advances in open surgery understanding.

3D reasoning is crucial in areas like robotics and autonomous driving. Due to the high cost of 3D data acquisition, some recent methods attempt to enable LLMs to perform 3D reasoning through multi-view images, thereby transferring the powerful 2D reasoning capabilities of LLMs to 3D environments. However, these methods face challenges: either they use redundant views that contain many perspectives irrelevant to the question, or they rely on globally aggregated multi-view representations, losing the fine-grained vision-language correlations. To tackle these challenges, we propose 3DMulti-LLM, which mainly consists of three components: a COT selector, a question-guided fusion block, and pre-trained LLMs. Specifically, first, the COT selector leverages the powerful chain-of-thought reasoning capabilities of LLMs to identify question-related multi-view images. In this way, 3DMulti-LLM can eliminate a substantial amount of interference from unnecessary viewpoints. Then, we propose a question-guided fusion block for integrating multi-view features via question-guided interaction among various viewpoints. Finally, the pre-trained LLMs are utilized to reason in 3D scenes directly through multi-view features. Notably, our approach understands the 3D scene solely through multi-view images, without requiring the input of point cloud information or additional 3D feature extraction. Through our experiments, 3DMulti-LLM achieves impressive performance and surpasses existing 3D-input-free methods by + 12.2% and + 7.1% on ScanQA and 3DMV-VQA datasets, respectively.

In real-world information retrieval scenarios, addressing data incompleteness is crucial for providing accurate similarity scores and ensuring reliable results for downstream tasks. Previous Similarity Matrix Completion (SMC) methods aim to estimate a similarity matrix that exhibits positive semi-definiteness (PSD) from an inaccurate one. However, these methods are inadequate in cases where the initial similarity matrix$S^{0}$exhibits PSD. In this paper, we propose a novel SMC framework that simultaneously explores the symmetric, positive semi-definiteness (PSD), and low-rank properties to provide an accurate and efficient solution for similarity search tasks. Specifically, we exploit the low-rank property and introduce an efficient specialized Cholesky factorization (CF) technique into the conventional SMC framework, which is implemented via a regularizer. It improves computational efficiency by learning a smaller factorized matrix instead of the entire similarity matrix. Moreover, to enhance the optimality guarantees of SMC, we introduce a novel lower-rank matrix property and design two corresponding regularizers. Building upon these meticulous designs, our novel SMC framework, for the first time, ensures both efficiency and accuracy with theoretical guarantees. Consequently, we propose two novel algorithms, i.e., SMCFN/SMCRN, to implement the SMC framework. Theoretical analysis verifies the effectiveness and fast convergence speed of SMCFN/SMCRN. Extensive experiments on five real-world datasets validate our theoretical analysis: SMCFN/SMCRN achieves up to 42% lower RMSE (e.g., 0.34 vs. 0.59 on ImageNet) and 15% higher Recall than state-of-the-art baselines, while being the most efficient among all baseline methods.

Unified detection of digital and physical attacks in facial recognition systems has become a focal point of research in recent years. However, current multi-modal methods typically ignore the intra-class and inter-class variability across different types of attacks, leading to degraded performance. To address this limitation, we propose MoAE-CR, a framework that effectively leverages class-aware information for improved attack detection. Our improvements manifest at two levels, i.e., the feature and loss level. At the feature level, we propose Mixture-of-Attack-Experts (MoAEs) to capture more subtle differences among various types of fake faces. At the loss level, we introduce Class Regularization (CR) through the Disentanglement Module (DM) and the Cluster Distillation Module (CDM). The DM enhances class separability by increasing the distance between the centers of live and fake face classes. However, center-to-center constraints alone are insufficient to ensure distinctive representations for individual features. Thus, we propose the CDM to further cluster features around their class centers while maintaining separation from other classes. Moreover, specific attacks that significantly deviate from common attack patterns are often overlooked. To address this issue, our distance calculation prioritizes more distant features. Extensive experiments on two unified physical-digital attack datasets demonstrate the state-of-the-art performance of the proposed method.

Removing reflection from a single image is challenging due to the absence of general reflection priors. Although existing methods incorporate extensive user guidance for satisfactory performance, they often lack the flexibility to adapt user guidance in different modalities, and dense user interactions further limit their practicality. To alleviate these problems, this paper presents FIRM, a novel framework for Flexible Interactive image Reflection reMoval with various forms of guidance, where users can provide sparse visual guidance (e.g., points, boxes, or strokes) or text descriptions for better reflection removal. Firstly, we design a novel user guidance conversion module (UGC) to transform different forms of guidance into unified contrastive masks. The contrastive masks provide explicit cues for identifying reflection and transmission layers in blended images. Secondly, we devise a contrastive mask-guided reflection removal network that comprises a newly proposed contrastive guidance interaction block (CGIB). This block leverages a unique cross-attention mechanism that merges contrastive masks with image features, allowing for precise layer separation. The proposed framework requires only 10% of the guidance time needed by previous interactive methods, which makes a step-change in flexibility. Extensive results on public real-world reflection removal datasets validate that our method demonstrates state-of-the-art reflection removal performance.

In radar-camera 3D object detection, the radar point clouds are sparse and noisy, which causes difficulties in fusing camera and radar modalities. To solve this, we introduce a novel query-based detection method named Radar-Camera Transformer (RCTrans). Specifically, we first design a Radar Dense Encoder to enrich the sparse valid radar tokens, and then concatenate them with the image tokens. By doing this, we can fully explore the 3D information of each interest region and reduce the interference of empty tokens during the fusing stage. We then design a Pruning Sequential Decoder to predict 3D boxes based on the obtained tokens and random initialized queries. To alleviate the effect of elevation ambiguity in radar point clouds, we gradually locate the position of the object via a sequential fusion structure. It helps to get more precise and flexible correspondences between tokens and queries. A pruning training strategy is adopted in the decoder, which can save much time during inference and inhibit queries from losing their distinctiveness. Extensive experiments on the large-scale nuScenes dataset prove the superiority of our method, and we also achieve new state-of-the-art radar-camera 3D detection results.

In recent years, diffusion models have revolutionized visual generation, outperforming traditional frameworks like Generative Adversarial Networks (GANs). However, generating images of humans with realistic semantic parts, such as hands and faces, remains a significant challenge due to their intricate structural complexity. To address this issue, we propose a novel post-processing solution named RealisHuman. The RealisHuman framework operates in two stages. First, it generates realistic human parts, such as hands or faces, using the original malformed parts as references, ensuring consistent details with the original image. Second, it seamlessly integrates the rectified human parts back into their corresponding positions by repainting the surrounding areas to ensure smooth and realistic blending. The RealisHuman framework significantly enhances the realism of human generation, as demonstrated by notable improvements in both qualitative and quantitative metrics.

Score Distillation Sampling (SDS) has been successfully extended to text-driven 3D scene editing with 2D pretrained diffusion models. However, SDS-based editing methods suffer from lengthy optimization processes with slow inference and low quality. We attribute the issue of lengthy optimization to the stochastic optimization scheme used in SDS-based editing, where many steps may conflict with each other (e.g., the inherent trade-off between editing and preservation). To reduce this internal conflict and speed up the editing process, we propose to separate editing and preservation in time with a diffusion time schedule and frame the 3D editing optimization process as a diffusion bridge sampling process. Motivated by the analysis above, we introduce DaCapo, a fast diffusion sampling-like 3D editing method that incorporates a novel stacked bridge framework, which estimates a direct diffusion bridge between source and target distribution with only a pretrained 2D diffusion model. Specifically, It models the editing process as a combination of inversion and generation, where both processes happen simultaneously as a stack of Diffusion Bridges. DaCapo shows a 15× speed-up with comparable results to the state-of-the-art SDS-based method. It completes the process in just 2,500 steps on a single GPU and accommodates a variety of 3D representation methods.

It is highly desirable to obtain a model that can generate high-quality 3D meshes from text prompts in just seconds. While recent attempts have adapted pre-trained text-to-image diffusion models, such as Stable Diffusion (SD), into generators of 3D representations (e.g., Triplane), they often suffer from poor quality due to the lack of sufficient high-quality 3D training data. Aiming at overcoming the data shortage, we propose a novel training scheme, termed as Progressive Rendering Distillation (PRD), eliminating the need for 3D ground-truths by distilling multi-view diffusion models and adapting SD into a native 3D generator. In each iteration of training, PRD uses the U-Net to progressively denoise the latent from random noise for a few steps, and in each step it decodes the denoised latent into 3D output. Multi-view diffusion models, including MVDream and RichDreamer, are used in joint with SD to distill text-consistent textures and geometries into the 3D outputs through score distillation. Since PRD supports training without 3D ground-truths, we can easily scale up the training data and improve generation quality for challenging text prompts with creative concepts. Meanwhile, PRD can accelerate the inference speed of the generation model in just a few steps. With PRD, we train a Triplane generator, namely TriplaneTurbo, which adds only 2.5% trainable parameters to adapt SD for Triplane generation. TriplaneTurbo outperforms previous text-to-3D generators in both efficiency and quality. Specifically, it can produce high-quality 3D meshes in 1.2 seconds and generalize well for challenging text input. The code is available at github.com/theEricMa/TriplaneTurbo.

To tackle the threat of fake news, the task of detecting and grounding multi-modal media manipulation (DGM4) has received increasing attention. However, most state-of-the-art methods fail to explore the fine-grained consistency within local content, usually resulting in an inadequate perception of detailed forgery and unreliable results. In this paper, we propose a novel approach named Contextual-Semantic Consistency Learning (CSCL) to enhance the fine-grained perception ability of forgery for DGM4. Two branches for image and text modalities are established, each of which contains two cascaded decoders, i.e., Contextual Consistency Decoder (CCD) and Semantic Consistency Decoder (SCD), to capture within-modality contextual consistency and across-modality semantic consistency, respectively. Both CCD and SCD adhere to the same criteria for capturing fine-grained forgery details. To be specific, each module first constructs consistency features by leveraging additional supervision from the heterogeneous information of each token pair. Then, the forgery-aware reasoning or aggregating is adopted to deeply seek forgery cues based on the consistency features. Extensive experiments on DGM4datasets prove that CSCL achieves new state-of-the-art performance, especially for the results of grounding manipulated content. Codes and weights are avaliable at https://github.com/liyih/CSCL.

Masked Image Modeling (MIM) with Vector Quantization (VQ) has achieved great success in both self-supervised pre-training and image generation. However, most existing methods struggle to address the trade-off in shared latent space for generation quality vs. representation learning and efficiency. To push the limits of this paradigm, we propose MergeVQ, which incorporates token merging techniques into VQ-based generative models to bridge the gap between image generation and visual representation learning in a unified architecture. During pre-training, MergeVQ decouples top-k semantics from latent space with the token merge module after self-attention blocks in the encoder for subsequent Look-up Free Quantization (LFQ) and global alignment and recovers their fine-grained details through cross-attention in the decoder for reconstruction. As for second-stage generation, we introduce MergeAR, which performs KV Cache compression for efficient raster-order prediction. Extensive experiments on ImageNet verify that MergeVQ as an AR generative model achieves competitive performance in both visual representation learning and image generation tasks while maintaining favorable token efficiency and inference speed. Code and model will be available at https://apexgen-x.github.io/MergeVQ.

Recent advancements in 3D object reconstruction have been remarkable, yet most current 3D models rely heavily on existing 3D datasets. The scarcity of diverse 3D datasets results in limited generalization capabilities of 3D reconstruction models. In this paper, we propose a novel framework for boosting 3D reconstruction with multi-view refinement (MVBoost) by generating pseudo-GT data. The key of MVBoost is combining the advantages of the high accuracy of the multi-view generation model and the consistency of the 3D reconstruction model to create a reliable data source. Specifically, given a single-view input image, we employ a multi-view diffusion model to generate multiple views, followed by a large 3D reconstruction model to produce consistent 3D data. MVBoost then adaptively refines these multi-view images, rendered from the consistent 3D data, to build a large-scale multi-view dataset for training a feed-forward 3D reconstruction model. Additionally, the input view optimization is designed to optimize the corresponding viewpoints based on the user’s input image, ensuring that the most important viewpoint is accurately tailored to the user’s needs. Extensive evaluations demonstrate that our method achieves superior reconstruction results and robust generalization compared to prior works.

Identifying multiple novel classes in an image, known as open-vocabulary multi-label recognition, is a challenging task in computer vision. Recent studies explore the transfer of powerful vision-language models such as CLIP. However, these approaches face two critical challenges: (1) The local semantics of CLIP are disrupted due to its global pre-training objectives, resulting in unreliable regional predictions. (2) The matching property between image regions and candidate labels has been neglected, relying instead on naive feature aggregation such as average pooling, which leads to spurious predictions from irrelevant regions. In this paper, we present RAM (Recover And Match), a novel framework that effectively addresses the above issues. To tackle the first problem, we propose Ladder Local Adapter (LLA) to enforce refocusing on local regions, recovering local semantics in a memory-friendly way. For the second issue, we propose Knowledge-Constrained Optimal Transport (KCOT) to suppress meaningless matching to non-GT labels by formulating the task as an optimal transport problem. As a result, RAM achieves state-of-the-art performance on various datasets from three distinct domains, and shows great potential to boost the existing methods. Code: https://github.com/EricTan7/RAM.

Test-time adaptation with pre-trained vision-language models, such as CLIP, aims to adapt the model to new, potentially out-of-distribution test data. Existing methods calculate the similarity between visual embedding and learnable class embeddings, which are initialized by text embeddings, for zero-shot image classification. In this work, we first analyze this process based on Bayes theorem, and observe that the core factors influencing the final prediction are the likelihood and the prior. However, existing methods essentially focus on adapting class embeddings to adapt likelihood, but they often ignore the importance of prior. To address this gap, we propose a novel approach, Bayesian Class Adaptation (BCA), which in addition to continuously updating class embeddings to adapt likelihood, also uses the posterior of incoming samples to continuously update the prior for each class embedding. This dual updating mechanism allows the model to better adapt to distribution shifts and achieve higher prediction accuracy. Our method not only surpasses existing approaches in terms of performance metrics but also maintains superior inference rates and memory usage, making it highly efficient and practical for real-world applications.

Transparent videos with alpha channels play a crucial role in film production, advertising, and augmented reality fields. However, there is currently no available method for producing transparent videos. Traditional methods are time-consuming and labor-intensive, and employing alternative approaches for this task will result in inaccurate transparent regions, constrained motion, and artifacts. To address these challenges, we propose Layer-Animate, the first method capable of generating transparent videos. Our method comprises two stages: in the first stage, transparent images are generated as the base images to provide content and transparency information for the next stage. In the second stage, Inter-Frame Attention is applied to decouple content from motion, enabling the motion module to focus better on action. Layer-Animate is the first method used to generate transparent videos with accurate transparent regions, sufficient motion, and no artifacts, as demonstrated by notable improvements in qualitative and quantitative metrics.

Large-scale text-to-image diffusion models have demonstrated impressive capabilities for downstream tasks by leveraging strong vision-language alignment from generative pre-training. Recently, a number of works have explored how to use the power of text-to-image diffusion models for text-image matching tasks. While previous generative text-image matching methods have shown potential in retrieving the most challenging candidates, they still suffer from extremely slow retrieval speeds and lack the ability to handle temporal dimensions, making them impractical for text-video retrieval. In this paper, we propose ZSGenRet, a simple yet effective zero-shot generative text-vision retrieval framework for both text-image and text-video tasks, based on pre-trained text-to-image diffusion models. We further incorporate inversion saliency detection to identify key frames in videos and enhance the semantic representation of the vision encoder. Experimental results demonstrate that ZSGenRet significantly improves text-video retrieval performance and achieves competitive results on text-image retrieval while remarkably improving efficiency. To the best of our knowledge, the proposed ZSGenRet is the first to explore zero-shot text-video retrieval based on diffusion models.

Speech-driven 3D facial animation is important for many multimedia applications. Recent work has shown promise in using either Diffusion models or Transformer architectures for this task. However, their mere aggregation does not lead to improved performance. We suspect this is due to a shortage of paired audio-4D data, which is crucial for the Transformer to effectively perform as a denoiser within the Diffusion framework. To tackle this issue, we present DiffSpeaker, a Transformer-based network equipped with novel biased conditional attention modules. These modules serve as substitutes for the traditional self/cross-attention in standard Transformers, incorporating thoughtfully designed biases that steer the attention mechanisms to concentrate on both the relevant task-specific and diffusion-related conditions. We also explore the trade-off between accurate lip synchronization and non-verbal facial expressions within the Diffusion paradigm. Experiments show our model achieves state-of-the-art performance on existing benchmarks, and fast inference speed owing to its ability to generate facial motions in parallel. Our code is avalable at https://github.com/theEricMa/DiffSpeaker.

Explainable artificial intelligence (XAI) aims to bring transparency to black-box neural networks. Many innovative explainable methods provide rich and multifaceted explanations. However, these methods often have complex mechanisms or a high complexity of explanations, making them difficult for people to understand. To address this challenge, we introduce a concept-based explainable method which can improve the readability of deep neural network explanations by obtaining a concise set of high-quality explanations. We reduce the explanation redundancy by weighted hierarchical clustering, thereby obtaining a set of explanations that completely describe the input image and are crucial to the network’s decision-making. Compared with existing XAI methods, our approach reduces the complexity while ensuring the integrity and comprehensibility of the explanation. In addition, we show that our approach can be traced back to the neuron-level explanation, which can also provide inspiration for model researchers to interpret the model. We validated the effectiveness of our approach through experiments in bird classification and facial recognition tasks. Specifically, we employed XAI to investigate the mechanisms behind face recognition, identifying critical neurons that correspond to key concepts in the recognition process.

3D whole-body mesh recovery aims to extract parameters for the human body, hands, and head from a single human image. Most applications related to human mesh recovery, such as physical fitness motion capture and operating room motion capture, necessitate real-time video stream processing. However, existing methods ignore the video processing and often require significant computational resources, making real-time performance unattainable and greatly limiting their practicality. Moreover, noticeable misalignments are often observed when concatenating them back to the body and reprojecting them onto the image. In this paper, we propose a streaming framework for whole-body mesh recovery in the video. First, we simplify pose regression by leveraging the root nodes of the hands and head to locate each component. Second, for temporal optimization, we incorporate attention mechanisms related to keypoint velocity to incorporate information from previous frames and achieve more stable and smooth motions. Finally, we propose a multi-view projection loss to eliminate the ambiguity caused by inaccurate 3D regression and pose estimation in computing reprojection errors. The combination of these enables our method to achieve real-time inference speed while maintaining accuracy and stability.

Face recognition (FR) systems are widely deployed but often struggle due to unconstrained image-capturing conditions. Face image quality assessment (FIQA), applied before recognition, mitigates these challenges by filtering out unreliable samples. Current leading FIQA methods evaluate image quality based on the characteristics observed within the FR model pipeline. However, they leave out the inherent differences in identity embeddings between high-and low-quality face images. To this end, we propose Gen-FIQA, which utilizes a generative model to probe and amplify this difference. Specifically, we extract the identity embedding from an input image using a pre-trained FR model, and then use it as a conditioning signal to generate several face images of the same identity. This generation process leverages the inherent prior in the generative model to translate the difference in identity embedding space back to pixel space. To quantify these differences, the quality score is computed as the average cosine similarity between embeddings from the original and generated images. To improve computational efficiency, we further distill GenFIQA into a lightweight regression-based variant, GenFIQA(R). Extensive experiments across five benchmark datasets and four FR models demonstrate the superiority of our methods over thirteen state-of-the-art FIQA methods.

Face forgery detection is a challenging problem due to the diversity and rapid iteration of face manipulation methods, especially in detecting unknown forgery types. To address this challenge, we explore the common features shared among various forgery types. We find that even though multiple manipulation methods leave different invisible forgery traces, the fake faces often exhibit a similar overall pattern of discomfort. Such discomfort can serve as a universal clue across multiple forgery types, thereby possessing the potential to achieve strong generalization in face forgery detection. To this end, we utilize Vision-Language Models (VLMs) to simulate the cognitive process from perceiving the image to generating the sense of discomfort and propose a multi-task Forgery-Discomfort Joint Learning (FDJL) framework to leverage VLMs to perceive and identify fake faces by integrating discomfort cues. Specifically, we collect a Facial Discomfort dataset guided by the uncanny valley theory, enabling the model to extract and learn discomfort features. Extensive experiments demonstrate that our method achieves state-of-the-art performance and exhibits the best generalization for unknown forgery types.

With the rapid growth usage of face recognition in people’s daily life, face anti-spoofing becomes increasingly important to avoid malicious attacks. Recent face anti-spoofing models can reach a high classification accuracy on multiple datasets but these models can only tell people "this face is fake" while lacking the explanation to answer "why it is fake". Such a system undermines trustworthiness and causes user confusion, as it denies their requests without providing any explanations. In this paper, we incorporate XAI into face anti-spoofing and propose a new problem termed X-FAS (eXplainable Face Anti-Spoofing) empowering face anti-spoofing models to provide an explanation. We propose SPTD (SPoof Trace Discovery), an X-FAS method which can discover spoof concepts and provide reliable explanations on the basis of discovered concepts. To evaluate the quality of X-FAS methods, we propose an X-FAS benchmark with annotated spoof traces by experts. We analyze SPTD explanations on face anti-spoofing dataset and compare SPTD quantitatively and qualitatively with previous XAI methods on proposed X-FAS benchmark. Experimental results demonstrate SPTD’s ability to generate reliable explanations.

Recently, significant advancements have been made in audio-driven talking face generation. While GAN-based methods are widely used in this task, they struggle to achieve simultaneous lip accuracy and high-fidelity. Generated lip shapes tend to be overly influenced by the lip of reference images that provide identity information, leading to unstable and unsynchronized results. Moreover, the synthesized face frequently suffers from blurred teeth, skin textures, and compromised facial identity. To address these challenges, we propose an effective and innovative training strategy that simultaneously ensures lip synchrony and facial fidelity. First, we adaptively select the reference image using a hard-mining based strategy to prevent the network from simply copying the reference lip, enhancing the stability and synchronicity of lip movements. Second, we incorporate high-resolution facial images in training a quality discriminator within the GAN loss, improving the generated faces’ fidelity. Third, a global-to-detail training strategy is employed, starting with strengthening synchrony and then image quality to preserve identity and visual details. Experiments on the HDTF dataset demonstrate that our method achieves state-of-the-art performance in both lip accuracy and image quality.

Surgical phase recognition is critical for assisting surgeons in understanding surgical videos. Existing studies focused more on online surgical phase recognition, by leveraging preceding frames to predict the current frame. Despite great progress, they formulated the task as a series of frame-wise classification, which resulted in a lack of global context of the entire procedure and incoherent predictions. Moreover, besides online analysis, accurate offline surgical phase recognition is also in significant clinical need for retrospective analysis, and existing online algorithms do not fully analyze the entire video, thereby limiting accuracy in offline analysis. To over-come these challenges and enhance both online and offline inference capabilities, we propose a universal Surgical Phase LocalizAtion Network, named SurgPLAN++, with the principle of temporal detection. To ensure a global understanding of the surgical procedure, we devise a phase localization strategy for SurgPLAN ++ to predict phase segments across the entire video through phase proposals. For online analysis, to generate high-quality phase proposals, SurgPLAN++ incorporates a data augmentation strategy to extend the streaming video into a pseudo-complete video through mirroring, center-duplication, and down-sampling. For offline analysis, SurgPLAN++ capi-talizes on its global phase prediction framework to continu-ously refine preceding predictions during each online inference step, thereby significantly improving the accuracy of phase recognition. We perform extensive experiments to validate the effectiveness, and our SurgPLAN++ achieves remarkable performance in both online and offline modes, which outper-forms state-of-the-art methods. The source code is available at https://github.com/franciszchenlSurgPLAN-Plus.

Humans' innate ability to decompose scenes into objects allows for efficient understanding, predicting, and planning. In light of this, Object-Centric Learning (OCL) attempts to endow networks with similar capabilities, learning to represent scenes with the composition of objects. However, existing OCL models only learn through reconstructing the input images, which does not assist the model in distinguishing objects, resulting in suboptimal object-centric representations. This flaw limits current object-centric models to relatively simple downstream tasks. To address this issue, we draw on humans’ top-down vision pathway and propose Top-Down Guided Network (TDGNet), which includes a top-down pathway to improve object-centric representations. During training, the top-down pathway constructs guidance with high-level object-centric representations to optimize low-level grid features output by the backbone. While during inference, it refines object-centric representations by detecting and solving conflicts between low- and high-level features. We show that TDGNet outperforms current object-centric models on multiple datasets of varying complexity. In addition, we expand the downstream task scope of object-centric representations by applying TDGNet to the field of robotics, validating its effectiveness in downstream tasks including video prediction and visual planning. Code will be available at https://github.com/zoujunhong/RHGNet.

The rapid evolution of deepfake generation technologies necessitates the development of robust face forgery detection algorithms. Recent studies have demonstrated that wavelet analysis can enhance the generalization abilities of forgery detectors. Wavelets effectively capture key facial contours, often slender, fine-grained, and globally distributed, that may conceal subtle forgery artifacts imperceptible in the spatial domain. However, current wavelet-based approaches fail to fully exploit the distinctive properties of wavelet data, resulting in sub-optimal feature extraction and limited performance gains. To address this challenge, we introduce WMamba, a novel wavelet-based feature extractor built upon the Mamba architecture. WMamba maximizes the utility of wavelet information through two key innovations. First, we propose Dynamic Contour Convolution (DCConv), which employs specially crafted deformable kernels to adaptively model slender facial contours. Second, by leveraging the Mamba architecture, our method captures long-range spatial relationships with linear complexity. This efficiency allows for the extraction of fine-grained, globally distributed forgery artifacts from small image patches. Extensive experiments show that WMamba achieves state-of-the-art (SOTA) performance, highlighting its effectiveness in face forgery detection.

The rise of realistic digital face generation and manipulation poses significant social risks. The primary challenge lies in the rapid and diverse evolution of generation techniques, which often outstrip the detection capabilities of existing models. To defend against the ever-evolving new types of forgery, we need to enable our model to quickly adapt to new domains with limited computation and data while avoiding forgetting previously learned forgery types. In this work, we posit that genuine facial samples are abundant and relatively stable in acquisition methods, while forgery faces continuously evolve with the iteration of manipulation techniques. Given the practical infeasibility of exhaustively collecting all forgery variants, we frame face forgery detection as a continual learning problem and allow the model to develop as new forgery types emerge. Specifically, we employ a Developmental Mixture of Experts (MoE) architecture that uses LoRA models as its individual experts. These experts are organized into two groups: a Real-LoRA to learn and refine knowledge of real faces, and multiple Fake-LoRAs to capture incremental information from different forgery types. To prevent catastrophic forgetting, we ensure that the learning direction of Fake-LoRAs is orthogonal to the established subspace. Moreover, we integrate orthogonal gradients into the orthogonal loss of Fake-LoRAs, preventing gradient interference throughout the training process of each task. Experimental results under both the datasets and manipulation types incremental protocols demonstrate the effectiveness of our method.

Metagenomic data, comprising mixed multi-species genomes, are prevalent in diverse environments like oceans and soils, significantly impacting human health and ecological functions. However, current research relies on K-mer, which limits the capture of structurally and functionally relevant gene contexts. Moreover, these approaches struggle with encoding biologically meaningful genes and fail to address the one-to-many and many-to-one relationships inherent in metagenomic data. To overcome these challenges, we introduce FGeneBERT, a novel metagenomic pre-trained model that employs a protein-based gene representation as a context-aware and structure-relevant tokenizer. FGeneBERT incorporates masked gene modeling to enhance the understanding of inter-gene contextual relationships and triplet enhanced metagenomic contrastive learning to elucidate gene sequence-function relationships. Pre-trained on over 100 million metagenomic sequences, FGeneBERT demonstrates superior performance on metagenomic datasets at four levels, spanning gene, functional, bacterial, and environmental levels and ranging from 1 to 213 k input sequences. Case studies of ATP synthase and gene operons highlight FGeneBERT's capability for functional recognition and its biological relevance in metagenomic research.

Spatial transcriptomics (ST) technologies have revolutionized the study of gene expression patterns in tissues by providing multimodal data, including transcriptomic (Tra.), spatial, and morphological modalities, thereby offering new opportunities to understand tissue biology beyond traditional Tra. However, we identify the modality bias phenomenon in ST data species, i.e. the inconsistent contribution of different modalities to the labels leads to a tendency for the analysis methods to retain the information of the dominant modality. How to mitigate the adverse effects of modality bias to satisfy various downstream tasks remains a fundamental challenge. This paper introduces Multiple-modality Structure Transformation, named MuST, a novel methodology to tackle the challenge. MuST integrates the multi-modality information contained in the ST data effectively into a uniform latent space to provide a foundation for all the downstream tasks. It learns intrinsic local structures by topology discovery strategy and topology fusion loss function to solve the inconsistencies among different modalities. Thus, these topology-based and deep learning techniques provide a solid foundation for a variety of analytical tasks while coordinating different modalities. The effectiveness of MuST is assessed by performance metrics and biological significance. The results show that it outperforms existing state-of-the-art methods with clear advantages in the precision of identifying and preserving structures of tissues and biomarkers. MuST offers a versatile toolkit for the intricate analysis of complex biological systems.

Face Anti-Spoofing (FAS) is essential for securing face recognition systems against presentation attacks. Recent advances in sensor technology and multimodal learning have enabled the development of multimodal FAS systems. However, existing methods often struggle to generalize to unseen attacks and diverse environments due to two key challenges: (1) Modality unreliability, where sensors such as depth and infrared suffer from severe domain shifts, impairing the reliability of cross-modal fusion; and (2) Modality imbalance, where over-reliance on a dominant modality weakens the model's robustness against attacks that affect other modalities. To overcome these issues, we propose MMDG++, a multimodal domain-generalized FAS framework built upon the vision-language model CLIP. In MMDG++, we design the Uncertainty-Guided Cross-Adapter++ (U-Adapter++) to filter out unreliable regions within each modality, enabling more reliable multimodal interactions. Additionally, we introduce Rebalanced Modality Gradient Modulation (ReGrad) for adaptive gradient modulation to balance modality convergence. To further enhance generalization, propose Asymmetric Domain Prompts (ADPs) that leverage CLIP's language priors to learn generalized decision boundaries across modalities. We also develop a novel multimodal FAS benchmark to evaluate generalizability under various deployment conditions. Extensive experiments across this benchmark show our method outperforms state-of-the-art FAS methods, demonstrating superior generalization capability.

Sign Language Representation Learning (SLRL) is crucial for a range of sign language-related downstream tasks such as Sign Language Translation (SLT) and Sign Language Retrieval (SLRet). Recently, many gloss-based and gloss-free SLRL methods have been proposed, showing promising performance. Among them, the gloss-free approach shows promise for strong scalability without relying on gloss annotations. However, it currently faces suboptimal solutions due to challenges in encoding the intricate, context-sensitive characteristics of sign language videos, mainly struggling to discern essential sign features using a non-monotonic video-text alignment strategy. Therefore, we introduce an innovative pretraining paradigm for gloss-free SLRL, called C2RL, in this paper. Specifically, rather than merely incorporating a non-monotonic semantic alignment of video and text to learn language-oriented sign features, we emphasize two pivotal aspects of SLRL: Implicit Content Learning (ICL) and Explicit Context Learning (ECL). ICL delves into the content of communication, capturing the nuances, emphasis, timing, and rhythm of the signs. In contrast, ECL focuses on understanding the contextual meaning of signs and converting them into equivalent sentences. Despite its simplicity, extensive experiments confirm that the joint optimization of ICL and ECL results in robust sign language representation and significant performance gains in gloss-free SLT and SLRet tasks. Notably, C2RL improves the BLEU-4 score by +5.3 on P14T, +10.6 on CSL-daily, +6.2 on OpenASL, and +1.3 on How2Sign. It also boosts the R@1 score by +8.3 on P14T, +14.4 on CSL-daily, and +5.9 on How2Sign. Additionally, we set a new baseline for the OpenASL dataset in the SLRet task.

As an emerging direction of machine learning, multi-target domain adaptation (MTDA) aims to address the challenges of adapting models to multiple target domains. However, existing studies often focus on single-target domain adaptation or fail to delve into the complexities associated with multiple target domains. So there is a notable lack of comprehensive research and exploration in MTDA. Consequently, we propose a cross-attention with conditional matching for MTDA that intends to overcome the challenges posed by domain discrepancy, multi-target domain heterogeneity, and scalability. Foremost, we design a novel multi-target conditional matching that aims to align the sample distribution by leveraging nearest neighbor principle. This strategy takes into account the unique characteristics of each target domain, facilitating adaptive adaptation across multiple domains. Furthermore, we use the transformer module and well-design a cross-attention mechanism to facilitate the alignment of distributions across the source and target domains, as well as among the target domains, thus mitigating discrepancies among multiple domains. Through integrating the cross-attention mechanism into the training phase, attaining effective alignment of cross-domain distributions, we improve the adaptability and performance of the method. By the end, our approach demonstrates effective and superior experimental results indicating the significance of our work.

Vision-language pretraining (VLP) models have demonstrated outstanding performance in image-text understanding tasks but remain highly susceptible to transferable adversarial attacks. While ensemble-based guided attacks improve adversarial transferability by increasing the diversity of image-text pairs, they primarily rely on spatial-domain data augmentation, which can lead to model overfitting to image details and limit the generalization capability of attacks. To address this limitation, this study proposes a frequency-domain adjustment-based adversarial attack method that modifies specific frequency components of input images to reduce detail interference and enhance the stability of adversarial examples. Additionally, a fine-grained feature extraction technique is introduced to optimize image-text alignment, further improving the transferability of cross-modal attacks. Experimental results demonstrate that the proposed method achieves superior attack transferability and generalization performance across two major VLP architectures, fusion models and alignment models, as well as multiple tasks on the Flickr30 K and MSCOCO datasets.

Facial recognition systems are vulnerable to physical (e.g., printed photos) and digital (e.g., DeepFake) face attacks. Existing methods struggle to simultaneously detect physical and digital attacks due to: 1) significant intra-class variations between these attack types, and 2) the inadequacy of spatial information alone to comprehensively capture live and fake cues. To address these issues, we propose a unified attack detection model termed Frequency-Aware and Attack-Agnostic CLIP (FA3-CLIP), which introduces attack-agnostic prompt learning to express generic live and fake cues derived from the fusion of spatial and frequency features, enabling unified detection of live faces and all categories of attacks. Specifically, the attack-agnostic prompt module generates generic live and fake prompts within the language branch to extract corresponding generic representations from both live and fake faces, guiding the model to learn a unified feature space for unified attack detection. Meanwhile, the module adaptively generates the live/fake conditional bias from the original spatial and frequency information to optimize the generic prompts accordingly, reducing the impact of intra-class variations. We further propose a dual-stream cues fusion framework in the vision branch, which leverages frequency information to complement subtle cues that are difficult to capture in the spatial domain. In addition, a frequency compression block is utilized in the frequency stream, which reduces redundancy in frequency features while preserving the diversity of crucial cues. We also establish new challenging protocols to facilitate unified face attack detection effectiveness. Experimental results on multiple benchmarks demonstrate that FA3-CLIP significantly improves performance, reducing ACER by over 1.2% on UniAttackData, and increasing AUC by more than 3% as well as reducing EER by over 4% on the JFSFDB dataset.

The goal of unsupervised domain adaptation person re-identification (UDA Reid) is to achieve feature space alignment between the source domain and the target domain, so that the Reid model can effectively match pedestrians in the target domain. Creating the transitional domain is an effective approach, but existing models often have difficulty synthesizing transitional domains with sufficiently public features. To tackle this challenge, we propose an innovative approach named feature fusion transitional domain (F2TD-Reid), which comprises two essential components: the dictionary fusion module (DFM) and the transitional domain attention module (TDAM). Among them, the DFM utilizes a feature fusion to extract and reconstruct pedestrian images from instances, focusing on capturing the essential visual elements within the images. For the TDAM, it further refines the feature extraction of instance points through an innovative weighted attention mechanism. These two modules optimize the generation process of scaling factors, thereby facilitating the transfer of knowledge between the source domain and the target domain. Through a series of comparative experiments, we verify the superiority of the F2TD-Reid method in solving UDA Reid. The code is available at https://github.com/1x-x/F2TD-Reid.

Contemporary deep face recognition techniques predominantly utilize the Softmax loss function, designed based on the similarities between sample features and class prototypes. These similarities can be categorized into four types: in-sample target similarity, in-sample non-target similarity, out-sample target similarity, and out-sample non-target similarity. When a sample feature from a specific class is designated as the anchor, the similarity between this sample and any class prototype is referred to as in-sample similarity. In contrast, the similarity between samples from other classes and any class prototype is known as out-sample similarity. The terms target and non-target indicate whether the sample and the class prototype used for similarity calculation belong to the same identity or not. The conventional Softmax loss function promotes higher in-sample target similarity than in-sample non-target similarity. However, it overlooks the relation between in-sample and out-sample similarity. In this paper, we propose Global Cross-Entropy loss (GCE), which promotes 1) greater in-sample target similarity over both the in-sample and out-sample non-target similarity, and 2) smaller in-sample non-target similarity to both in-sample and out-sample target similarity. In addition, we propose to establish a bilateral margin penalty for both in-sample target and non-target similarity, so that the discrimination and generalization of the deep face model are improved. To bridge the gap between training and testing of face recognition, we adapt the GCE loss into a pairwise framework by randomly replacing some class prototypes with sample features. We designate the model trained with the proposed Global Cross-Entropy loss as GFace. Extensive experiments on several public face benchmarks, including LFW, CALFW, CPLFW, CFP-FP, AgeDB, IJB-C, IJB-B, MFR-Ongoing, and MegaFace, demonstrate the superiority of GFace over other methods. Additionally, GFace exhibits robust performance in general visual recognition task.

Pedestrian attribute recognition has achieved high accuracy by exploring the relations between image regions and attributes. However, existing methods typically adopt features directly extracted from the backbone or utilize a single structure (e.g., transformer) to explore the relations, leading to inefficient and incomplete relation mining. To overcome these limitations, this paper proposes a comprehensive relationship framework called Vision Transformer with Relation Exploration (ViT-RE) for pedestrian attribute recognition, which includes two novel modules, namely Attribute and Contextual Feature Projection (ACFP) and Relation Exploration Module (REM). In ACFP, attribute-specific features and contextual-aware features are learned individually to capture discriminative information tailored for attributes and image regions, respectively. Then, REM employs Graph Convolutional Network (GCN) Blocks and Transformer Blocks to concurrently explore attribute, contextual, and attribute-contextual relations. To enable fine-grained relation mining, a Dynamic Adjacency Module (DAM) is further proposed to construct instance-wise adjacency matrix for the GCN Block. Equipped with comprehensive relation information, ViT-RE achieves promising performance on three popular benchmarks, including PETA, RAP, and PA-100 K datasets. Moreover, ViT-RE achieves the first place in theWACV 2023 UPAR Challenge.

Unsupervised fault detection in multivariate time series (MTS) plays a vital role in ensuring the stable operation of complex systems. Traditional methods often assume that normal data follow a single Gaussian distribution and identify anomalies as deviations from this distribution. However, this simplified assumption fails to capture the diversity and structural complexity of real-world time series, which can lead to misjudgments and reduced detection performance in practical applications. To address this issue, we propose a new method that combines a neighborhood-driven data augmentation strategy with a multimanifold representation learning framework. By incorporating information from local neighborhoods, the augmentation module can simulate contextual variations of normal data, enhancing the model's adaptability to distributional changes. In addition, we design a structure-aware feature learning approach that encourages natural clustering of similar patterns in the feature space while maintaining sufficient distinction between different operational states. Extensive experiments on several public benchmark datasets demonstrate that our method achieves superior performance in terms of both accuracy and robustness, showing strong potential for generalization and real-world deployment.

The deep reinforcement learning (DRL) has made significant progress in various simulation environments. However, applying DRL methods to real-world scenarios poses certain challenges due to limitations in visual fidelity, scene complexity, and task diversity within existing environments. To address limitations and explore the potential ability of DRL, we developed a 3-D open-world first-person shooter (FPS) game called Unreal BattleGround (UBG) using the unreal engine (UE). UBG provides a realistic 3-D environment with variable complexity, random scenes, diverse tasks, and multiple scene interaction methods. This benchmark involves far more complex state-action spaces than classic pseudo-3-D FPS games (e.g., ViZDoom), making it challenging for DRL to learn human-level decision sequences. Then, we propose the object-aware hierarchically proximal policy optimization (OaH-PPO) method in the UBG. It involves a two-level hierarchy, where the high-level controller is tasked with learning option control, and the low-level workers focus on mastering subtasks. To boost the learning of subtasks, we propose three modules: an object-aware module for extracting depth detection information from the environment, potential-based intrinsic reward shaping for efficient exploration, and annealing imitation learning (IL) to guide the initialization. Experimental results have demonstrated the broad applicability of the UBG and the effectiveness of the OaH-PPO. We will release the code of the UBG and OaH-PPO after publication.

Open-set domain adaptation (OSDA) seeks to transfer knowledge from a labeled source domain to an unlabeled target domain containing novel classes. Traditional OSDA methods rarely account for the uncertainty in predictions and typically require additional training overhead. Evidential deep learning (EDL) transforms the model's predictions from point estimates to distributions over the probability simplex by replacing the standard softmax output of classification neural networks with Dirichlet distributions. Considering the presence of out-of-distribution novel classes in OSDA and the additional overhead of existing methods, we propose EDL for open-set active domain adaptation (EOSADA). Leveraging EDL, we construct an open-set classifier and employ a two-round selection strategy guided by the data uncertainty of target domain samples and semantic similarity scores with known classes. This strategy balances the selection of samples from known and novel classes while identifying informative samples, thereby maximizing the performance of the model in OSDA scenarios without modifying the model structure and utilizing a limited annotation budget. Extensive experiments demonstrate the superiority of our approach.

Recently, unsupervised cross-dataset person reidentification (Re-ID) has attracted more and more attention, which aims to transfer knowledge of a labeled source domain to an unlabeled target domain. There are two common frameworks: one is pixel-alignment of transferring low-level knowledge, and the other is feature-alignment of transferring high-level knowledge. In this article, we propose a novel recurrent autoencoder (RAE) framework to unify these two kinds of methods and inherit their merits. Specifically, the proposed RAE includes three modules, i.e., a feature-transfer (FT) module, a pixel-transfer (PT) module, and a fusion module. The FT module utilizes an encoder to map source and target images to a shared feature space. In the space, not only features are identity-discriminative but also the gap between source and target features is reduced. The PT module takes a decoder to reconstruct original images with its features. Here, we hope that the images reconstructed from target features are in the source style. Thus, the low-level knowledge can be propagated to the target domain. After transferring both high-and low-level knowledge with the two proposed modules above, we design another bilinear pooling layer to fuse both kinds of knowledge. Extensive experiments on Market-1501, DukeMTMC-ReID, and MSMT17 datasets show that our method significantly outperforms either pixel-alignment or feature-alignment Re-ID methods and achieves new state-of-the-art results.

To achieve accurate and robust object detection in the real-world scenario, various forms of images are incorporated, such as color, thermal, and depth. However, multimodal data often suffer from the position shift problem, i.e., the image pair is not strictly aligned, making one object has different positions in different modalities. For the deep learning method, this problem makes it difficult to fuse multimodal features and puzzles the convolutional neural network (CNN) training. In this article, we propose a general multimodal detector named aligned region CNN (AR-CNN) to tackle the position shift problem. First, a region feature (RF) alignment module with adjacent similarity constraint is designed to consistently predict the position shift between two modalities and adaptively align the cross-modal RFs. Second, we propose a novel region of interest (RoI) jitter strategy to improve the robustness to unexpected shift patterns. Third, we present a new multimodal feature fusion method that selects the more reliable feature and suppresses the less useful one via feature reweighting. In addition, by locating bounding boxes in both modalities and building their relationships, we provide novel multimodal labeling named KAIST-Paired. Extensive experiments on 2-D and 3-D object detection, RGB-T, and RGB-D datasets demonstrate the effectiveness and robustness of our method.

Vision-Language Models (VLMs) such as CLIP have demonstrated remarkable generalization capabilities to downstream tasks. However, existing prompt tuning based frameworks need to parallelize learnable textual inputs for all categories, suffering from massive GPU memory consumption when there is a large number of categories in the target dataset. Moreover, previous works require to include category names within prompts, exhibiting subpar performance when dealing with ambiguous category names. To address these shortcomings, we propose Compound Text-Guided Prompt Tuning (TGP-T) that significantly reduces resource demand while achieving superior performance. We introduce text supervision to the optimization of prompts, which enables two benefits: 1) releasing the model reliance on the pre-defined category names during inference, thereby enabling more flexible prompt generation; 2) reducing the number of inputs to the text encoder, which decreases GPU memory consumption significantly. Specifically, we found that compound text supervisions, i.e., category-wise and content-wise, is highly effective, since they provide inter-class separability and capture intra-class variations, respectively. Moreover, we condition the prompt generation on visual features through a module called Bonder, which facilitates the alignment between prompts and visual features. Extensive experiments on few-shot recognition and domain generalization demonstrate that TGP-T achieves superior performance with consistently lower training costs. It reduces GPU memory usage by 93% and attains a 2.5% performance gain on 16-shot ImageNet. The code is available at https://github.com/EricTan7/TGP-T.

Inversion methods, such as Textual Inversion, generate personalized images by incorporating concepts of interest provided by user images. However, existing methods often suffer from overfitting issues, where the dominant presence of inverted concepts leads to the absence of other desired concepts. It stems from the fact that during inversion, the irrelevant semantics in the user images are also encoded, forcing the inverted concepts to occupy locations far from the core distribution in the embedding space. To address this issue, we propose a method that guides the inversion process towards the core distribution for compositional embeddings. Additionally, we introduce a spatial regularization approach to balance the attention on the concepts being composed. Our method is designed as a post-training approach and can be seamlessly integrated with other inversion methods. Experimental results demonstrate the effectiveness of our proposed approach in mitigating the overfitting problem and generating more diverse and balanced compositions of concepts in the synthesized images. The source code is available at https://github.com/zhangxulu1996/Compositional-Inversion.

The surgical intervention is crucial to patient healthcare, and many studies have developed advanced algorithms to provide understanding and decision-making assistance for surgeons. Despite great progress, these algorithms are developed for a single specific task and scenario, and in practice require the manual combination of different functions, thus limiting the applicability. Thus, an intelligent surgical assistant is expected to accurately understand the surgeon’s intentions and accordingly conduct the specific tasks to support the surgical process. In this work, by improving advanced multimodal large language models (MLLMs), we propose a multimodal Surgical Function Calling (SurgFC) framework that can accurately understand the surgeon’s intention and complete a series of surgical understanding tasks, e.g., surgical scene analysis, surgical instrument detection, and segmentation on demand. Specifically, to achieve superior surgical multimodal understanding, we devise a mixture-of-projectors (MOP) module to align the surgical MLLM in SurgFC to balance the natural and surgical knowledge. Moreover, we devise a surgical Function-Calling Tuning strategy to enable the SurgFC to understand surgical intentions, and thus make a series of surgical function calls on demand to meet the needs of the surgeons. Extensive experiments on neurosurgery data confirm that our SurgFC can understand the surgeon’s intention more accurately than the existing MLLM, resulting in overwhelming performance in textual analysis and visual tasks. The source code is available at https://github.com/franciszchen/SurgFC.

Spatial transcriptomics (ST) technologies have revolutionized the study of gene expression patterns in tissues by providing multimodality data in transcriptomic, spatial, and morphological, offering opportunities for understanding tissue biology beyond transcriptomics. However, we identify the modality bias phenomenon in ST data species, i.e., the inconsistent contribution of different modalities to the labels leads to a tendency for the analysis methods to retain the information of the dominant modality. How to mitigate the adverse effects of modality bias to satisfy various downstream tasks remains a fundamental challenge. This paper introduces Multiple-modality Structure Transformation, named MuST, a novel methodology to tackle the challenge. MuST integrates the multi-modality information contained in the ST data effectively into a uniform latent space to provide a foundation for all the downstream tasks. It learns intrinsic local structures by topology discovery strategy and topology fusion loss function to solve the inconsistencies among different modalities. Thus, these topology-based and deep learning techniques provide a solid foundation for a variety of analytical tasks while coordinating different modalities. The effectiveness of MuST is assessed by performance metrics and biological significance. The results show that it outperforms existing state-of-the-art methods with clear advantages in the precision of identifying and preserving structures of tissues and biomarkers. MuST offers a versatile toolkit for the intricate analysis of complex biological systems. The code is available at https://github.com/zangzelin/code_Must.

Previous Sign Language Translation (SLT) methods achieve superior performance by relying on gloss annotations. However, labeling high-quality glosses is a labor-intensive task, which limits the further development of SLT. Although some approaches work towards gloss-free SLT through jointly training the visual encoder and translation network, these efforts still suffer from poor performance and inefficient use of the powerful Large Language Model (LLM). Most seriously, we find that directly introducing LLM into SLT will lead to insufficient learning of visual representations as LLM dominates the learning curve. To address these problems, we propose Factorized Learning assisted with Large Language Model (FLa-LLM) for gloss-free SLT. Concretely, we factorize the training process into two stages. In the visual initialing stage, we employ a lightweight translation model after the visual encoder to pre-train the visual encoder. In the LLM fine-tuning stage, we freeze the acquired knowledge in the visual encoder and integrate it with a pre-trained LLM to inspire the LLM’s translation potential. This factorized training strategy proves to be highly effective as evidenced by significant improvements achieved across three SLT datasets which are all conducted under the gloss-free setting.

Domain generalization (DG) based Face Anti-Spoofing (FAS) aims to improve the model's performance on unseen domains. Existing methods either rely on domain labels to align domain-invariant feature spaces, or disentangle generalizable features from the whole sample, which inevitably lead to the distortion of semantic feature structures and achieve limited generalization. In this work, we make use of large-scale VLMs like CLIP and leverage the textual feature to dynamically adjust the classifier's weights for exploring generalizable visual features. Specifically, we propose a novel Class Free Prompt Learning (CFPL) paradigm for DG FAS, which utilizes two lightweight transformers, namely Content Q-Former (CQF) and Style Q-Former (SQF), to learn the different semantic prompts conditioned on content and style features by using a set of learnable query vectors, respectively. Thus, the generalizable prompt can be learned by two improvements: (1) A Prompt-Text Matched (PTM) supervision is introduced to ensure CQF learns visual representation that is most informative of the content description. (2) A Diversified Style Prompt (DSP) technology is proposed to diversify the learning of style prompts by mixing feature statistics between instance-specific styles. Finally, the learned text features modulate visual features to generalization through the designed Prompt Modulation (PM). Extensive experiments show that the CFPL is effective and outperforms the state-of-the-art methods on several cross-domain datasets.

3D Morphable Models (3DMMs) provide promising 3D face reconstructions in various applications. However, existing methods struggle to reconstruct faces with extreme expressions due to deficiencies in supervisory signals, such as sparse or inaccurate landmarks. Segmentation information contains effective geometric contexts for face reconstruction. Certain attempts intuitively depend on differentiable renderers to compare the rendered silhouettes of reconstruction with segmentation, which is prone to issues like local optima and gradient instability. In this paper, we fully utilize the facial part segmentation geometry by introducing Part Re-projection Distance Loss (PRDL). Specifically, PRDL transforms facial part segmentation into 2D points and re-projects the reconstruction onto the image plane. Subsequently, by introducing grid anchors and computing different statistical distances from these anchors to the point sets, PRDL establishes geometry descriptors to optimize the distribution of the point sets for face reconstruction. PRDL exhibits a clear gradient compared to the renderer-based methods and presents state-of-the-art reconstruction performance in extensive quantitative and qualitative experiments. Our project is available at https://github.com/wang-zidu/3DDFA-V3.

Zhanyue Qin, Haochuan Wang, Deyuan Liu, Ziyang Song, Cunhang Fan, Zhao Lv, Jinlin Wu, Zhen Lei, Zhiying Tu, Dianhui Chu, Xiaoyan Yu, Dianbo Sui. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.

Face anti-spoofing is essential for ensuring the security of facial recognition systems against spoofing attacks. Recent methods have transferred Vision-Language models to face anti-spoofing (e.g., FLIP and CLIPC8), demonstrating that learning perception from supervision in natural language can enhance the model’s detection performance. However, such methods exhibit limited depth in the interaction between images and texts, resulting in poor performance on fine-grained understanding tasks such as face anti-spoofing. Besides, the lack of diversity in image-text pairs for face anti-spoofing further hinders such methods from playing their best. To address these issues, we propose a novel fine-tuning strategy for Vision-Language models in face anti-spoofing. This strategy introduces the Bootstrapping Language-Image Pre-training model (BLIP), known for its novel interaction mechanisms and superior image-text comprehension, to construct a more generalized feature representation for face anti-spoofing. Furthermore, we propose an Adapter module for the text branch to reduce the negative impact of insufficient data diversity and catastrophic forgetting. Extensive experiments conducted on various cross-domain testing benchmarks demonstrate the significant superiority of our method over the state-of-the-art, highlighting its effectiveness and robustness.

Face anti-spoofing (FAS) is pivotal in safeguarding the integrity of face recognition systems. Flexible-modal FAS utilizes multi-modal data and trains a unified model adaptable to any single-modal testing scenario. This innovation addresses the shortcomings of conventional multi-modal FAS approaches, which typically demand separate model training and deployment for each modality. However, existing flexible-modal FAS approaches activate specific network branches based on the modality of the tested sample. This not only increases the model’s parameters but also necessitates the provision of the image’s modality for testing, thereby constraining deployment flexibility. To address the issue, we present Compound Prompt Learning CLIP (CPL-CLIP), a novel method for flexible-modal FAS. This approach capitalizes on a learned textual prompt that is nearly independent of modality, thus bolstering class-based classification across arbitrary modalities. Specifically, our CPL-CLIP introduces a Dual-Branch Prompt (DBP), consisting of class and modal prompts that describe and guide classification, where each prompt is composed of learnable vectors and fixed templates. To further render the class prompt as modality-agnostic as possible, a Cosine Similarity Loss (CSL) is proposed to facilitate the maximal separation of the class prompt from the modality prompt. With only the class prompt utilized during testing, CPL-CLIP enables deployment in diverse modal testing scenarios without the necessity of the test image’s modality to be known. Extensive experiments demonstrate CPL-CLIP’s superiority over existing methods on several flexible-modal FAS benchmarks.

Facial recognition systems are susceptible to both physical and digital attacks, posing significant security risks. Traditional approaches often treat these two attack types separately due to their distinct characteristics. Thus, when being combined attacked, almost all methods could not deal. Some studies attempt to combine the sparse data from both types of attacks into a single dataset and try to find a common feature space, which is often impractical due to the space is difficult to be found or even non-existent. To overcome these challenges, we propose a novel approach that uses the sparse model to handle sparse data, utilizing different parameter groups to process distinct regions of the sparse feature space. Specifically, we employ the Mixture of Experts (MoE) framework in our model, expert parameters are matched to tokens with varying weights during training and adaptively activated during testing. However, the traditional MoE struggles with the complex and irregular classification boundaries of this problem. Thus, we introduce a flexible self-adapting weighting mechanism, enabling the model to better fit and adapt. In this paper, we proposed La-SoftMoE CLIP, which allows for more flexible adaptation to the Unified Attack Detection (UAD) task, significantly enhancing the model’s capability to handle diversity attacks. Experiment results demonstrate that our proposed method has SOTA performance.

AI-assisted operating room scene understanding is essential for the next generation of surgical interventions. Surgical instrument localization plays an important role in this context. However, existing instrument localization methods primarily focus on surgical instrument localization in endoscopy images and struggle with occlusions in broader operating room scenarios. In this work, we propose a weakly supervised instance segmentation framework, Pixel-driven Weakly-supervised Instance Segmentation (PWISeg), to solve the occluded instrument localization with low-cost annotations. Specifically, We utilize the projection relationship between the bounding box and the surgical instrument mask as a supervision signal to train PWISeg to predict coarse masks of surgical instruments. Then, we use the annotation of a few pixel points to train PWISeg to predict accurate masks of surgical instruments. To extensively validate the effectiveness, we collect and release a high-quality dataset, Surg-Inst that covers real-world hard cases of overlapping, dense placement, and various levels of instrument occlusion. Experiments demonstrate that our PWISeg achieves a remarkable performance advantage over state-of-the-art methods on both Surg-Inst and public HOSPI-Tools datasets.

In this work, borrowing a solution from the large-scale vision-language models (VLMs) instead of directly removing modality-specific signals from visual features, we propose a novel Flexible Modal CLIP (FM-CLIP) for flexible modal FAS, that can utilize text features to dynamically adjust visual features to be modality independent. In the visual branch, considering the huge visual differences of the same attack in different modalities, which makes it difficult for classifiers to flexibly identify subtle spoofing clues in different test modalities, we propose Cross-Modal Spoofing Enhancer (CMS-Enhancer). It includes a Frequency Extractor (FE) and Cross-Modal Interactor (CMI), aiming to map different modal attacks in a shared frequency space to reduce interference from modality-specific signals and enhance spoofing clues by leveraging cross-modal learning from the shared frequency space. In the text branch, we introduce a Language-Guided Patch Alignment (LGPA) based on prompt learning, which further guides the image encoder to focus on patch-level spoofing representations through dynamic weighting by text features. Thus, our FM-CLIP can flexibly test different modal samples by identifying and enhancing modality-agnostic spoofing cues. Finally, extensive experiments show that FM-CLIP is effective and outperforms state-of-the-art methods on multiple multi-modal datasets.

3D textured face reconstruction from sketches applicable in many scenarios such as animation, 3D avatars, artistic design, missing people search, etc., is a highly promising but underdeveloped research topic.On the one hand, the stylistic diversity of sketches leads to existing sketch-to-3D-face methods only being able to handle pose-limited and realistically shaded sketches.On the other hand, texture plays a vital role in representing facial appearance, yet sketches lack this information, necessitating additional texture control in the reconstruction process.This paper proposes a novel method for reconstructing controllable textured and detailed 3D faces from sketches, named S2TD-Face.S2TD-Face introduces a two-stage geometry reconstruction framework that directly reconstructs detailed geometry from the input sketch.To keep geometry consistent with the delicate strokes of the sketch, we propose a novel sketch-to-geometry loss that ensures the reconstruction accurately fits the input features like dimples and wrinkles.Our training strategies do not rely on hard-to-obtain 3D face scanning data or labor-intensive hand-drawn sketches.Furthermore, S2TD-Face introduces a texture control module utilizing text prompts to select the most suitable textures from a library and seamlessly integrate them into the geometry, resulting in a 3D detailed face with controllable texture.S2TD-Face surpasses existing state-of-the-art methods in extensive quantitative and qualitative experiments.Our project is available at https://github.com/wang-zidu/S2TD-Face.

In the mobile internet era, short videos are inundating people's lives. However, research on visual language models specifically designed for short videos has not yet received sufficient attention. Short videos are not just videos of limited duration. The prominent visual details and high information density of short videos differentiate them to long videos. In this paper, we propose the SpatioTemporal Fine-grained Description (STFVD) emphasizing on the uniqueness of short videos, which entails capturing the intricate details of the main subject and fine-grained movements. To this end, we create a comprehensive Short Video Advertisements Description (SVAD) dataset, comprising 34,930 clips from 5,046 videos. The dataset covers a range of topics, including 191 sub-industries, 649 popular products, and 470 trending games. Various efforts have been made in the data annotation process to ensure the inclusion of fine-grained spatiotemporal information, resulting in 34,930 high-quality annotations. Compared to existing datasets, samples in SVAD exhibit a superior text information density, suggesting that SVAD is more appropriate for the analysis of short videos. Based on the SVAD dataset, we develop a visual language model (SVAD-VLM) to generate spatiotemporal fine-grained description for short videos. We use a prompt-guided keyword generation task to efficiently learn key visual information. Moreover, we also utilize dual visual alignment to exploit the advantage of mixed-datasets training. Experiments on SVAD dataset demonstrate the challenge of STFVD and the competitive performance of proposed method compared to previous ones.

This paper presents a pilot study that explores the application of active learning, traditionally studied in the context of discriminative models, to generative models. We specifically focus on image synthesis personalization tasks. The primary challenge in conducting active learning on generative models lies in the open-ended nature of querying, which differs from the closed form of querying in discriminative models that typically target a single concept. We introduce the concept of anchor directions to transform the querying process into a semi-open problem. We propose a direction-based uncertainty sampling strategy to enable generative active learning and tackle the exploitation-exploration dilemma. Extensive experiments are conducted to validate the effectiveness of our approach, demonstrating that an open-source model can achieve superior performance compared to closed-source models developed by large companies, such as Google's StyleDrop. The source code is available at https://github.com/zhangxulu1996/GAL4Personalization.

Serialization-based methods, which serialize the 3D voxels and group them into multiple sequences before inputting to Transformers, have demonstrated their effectiveness in 3D object detection. However, serializing 3D voxels into 1D sequences will inevitably sacrifice the voxel spatial proximity. Such an issue is hard to be addressed by enlarging the group size with existing serialization-based methods due to the quadratic complexity of Transformers with feature sizes. Inspired by the recent advances of state space models (SSMs), we present a Voxel SSM, termed as Voxel Mamba, which employs a group-free strategy to serialize the whole space of voxels into a single sequence. The linear complexity of SSMs encourages our group-free design, alleviating the loss of spatial proximity of voxels. To further enhance the spatial proximity, we propose a Dual-scale SSM Block to establish a hierarchical structure, enabling a larger receptive field in the 1D serialization curve, as well as more complete local regions in 3D space. Moreover, we implicitly apply window partition under the group-free framework by positional encoding, which further enhances spatial proximity by encoding voxel positional information. Our experiments on Waymo Open Dataset and nuScenes dataset show that Voxel Mamba not only achieves higher accuracy than state-of-the-art methods, but also demonstrates significant advantages in computational efficiency. The source code is available at https://github.com/gwenzhang/Voxel-Mamba.

Source-free unsupervised domain adaptation (SFUDA) aims to conduct prediction on the target domain by leveraging knowledge from the well-trained source model. Due to the absence of source data in the SFUDA setting, the existing methods mainly build the target classifier by fine-tuning the source model incorporated with empirical adaptation losses. Although these methods have achieved somewhat promising results, nearly all of them typically suffer from the closed-fitting dilemma that their models are dominantly affected by these easy-to-distinguish instances than those hard-to-distinguish ones, resulting from the absence of the labeled source data. To address aforementioned issues, we propose the Dipolar Confidence Learning (DCL) for SFUDA. Specifically, we conduct positive confidence learning on the samples with standard outputs to avoid overfitting of the model to these samples. In contrast, we perform negative confidence learning for the samples with abnormal outputs to optimize the complementary label, which forces the network to pay more attention to these confusing samples. Furthermore, to achieve more generalized domain alignment, both the confidence-based fuzzy mixup and rotation-based self-supervised learning are respectively constructed to boost the representation ability of the target model. Finally, extensive experiments are conducted to demonstrate the effectiveness and performance superiority of the proposed method.

Remote sensing image fusion aims to generate a high-resolution multi/hyperspectral image by combining a high-resolution image with limited spectral data and a low-resolution image rich in spectral information. Current deep learning (DL) methods typically employ convolutional neural networks (CNNs) or Transformers for feature extraction and information integration. While CNNs are efficient, their limited receptive fields restrict their ability to capture global context. Transformers excel at learning global information but are computationally expensive. Recent advancements in the state space model (SSM), particularly Mamba, present a promising alternative by enabling global perception with low complexity. However, the potential of SSM for information integration remains largely unexplored. Therefore, we propose FusionMamba, an innovative method for efficient remote sensing image fusion. Our contributions are twofold. First, to effectively merge spatial and spectral features, we expand the single-input Mamba block to accommodate dual inputs, creating the FusionMamba block, which serves as a plug-and-play solution for information integration. Second, we incorporate Mamba and FusionMamba blocks into an interpretable network architecture tailored for remote sensing image fusion. Our designs utilize two U-shaped network branches, each primarily composed of four-directional (FD) Mamba blocks, to extract spatial and spectral features separately and hierarchically. The resulting feature maps are sufficiently merged in an auxiliary network branch constructed with FusionMamba blocks. Furthermore, we improve the representation of spectral information through an enhanced channel attention module. Quantitative and qualitative valuation results across six datasets demonstrate that our method achieves the state-of-the-art (SOTA) performance, underscoring the effectiveness of FusionMamba. The code is available athttps://github.com/PSRben/FusionMamba.

Face Anti-spoofing (FAS) is essential to secure face recognition systems from various physical attacks. However, recent research generally focuses on short-distance applications (i.e., phone unlocking) while lacking consideration of long-distance scenes (i.e., surveillance security checks). In order to promote relevant research and fill this gap in the community, we collect a large-scale Su rveillance Hi gh-Fi delity Mask (SuHiFiMask) dataset captured under 40 surveillance scenes, which has 101 subjects from different age groups with$232~3\text{D}$attacks (high-fidelity masks),$200~2\text{D}$attacks (posters, portraits, and screens), and 2 adversarial attacks. In this scene, low image resolution and noise interference are new challenges faced in surveillance FAS. Together with the SuHiFiMask dataset, we propose a Contrastive Quality-Invariance Learning (CQIL) network to alleviate the performance degradation caused by image quality from three aspects: 1) An Image Quality Variable module (IQV) is introduced to recover image information associated with discrimination by combining the super-resolution network. 2) Using generated sample pairs to simulate quality variance distributions to help contrastive learning strategies obtain robust feature representation under quality variation. 3) A Separate Quality Network (SQN) is designed to learn discriminative features independent of image quality. Finally, a large number of experiments verify the quality of the SuHiFiMask dataset and the superiority of the proposed CQIL.

The COVID-19 pandemic makes wearing masks mandatory. Existing CNN-based face recognition (FR) systems suffer from severe performance degradation as masks occlude the vital facial regions. Recently, Vision Transformers have shown promising performance in various vision tasks with quadratic computation costs. Swin Transformer first proposes a successive window attention mechanism allowing the cross-window connection and more computational efficiency. Despite its potential, the deployment of Swin Transformer in masked face recognition encounters two challenges: 1) the attention range is insufficient to capture locally compatible face regions. 2) Masked face recognition can be defined as an occlusion-robust classification task with a known occlusion position, i.e., the position of the mask is minor-varying, which is overlooked but efficient in improving the model’s recognition accuracy. To alleviate the above problem, we propose a Masked Face Transformer (MFT) with Masked Face-compatible Attention (MFA). The proposed MFA 1) introduces two additional window partition configurations, e.g., row shift and column shift, to enlarge the attention range in Swin with invariant computation costs, and 2) suppresses the interaction between the masked and non-masked regions to retain their discrepancies. Additionally, as mask occlusion leads to a separation between the masked and non-masked samples of the same identity, we propose to explore the relationship between them by a ClassFormer module to enhance intra-class aggregation. Extensive experiments show that MFT outperforms state-of-the-art masked face recognition methods in both simulated and real masked face testing datasets.

In this paper, we focus on estimating six degrees of freedom (6DoF) pose of a face from a single RGB image, which is an important but under-investigated problem in 3D face applications such as face reconstruction, forgery detection and virtual try-on. This problem is different from traditional face pose estimation and 3D face reconstruction since the distance from camera to face should be estimated, which can not be directly regressed due to the non-linearity of the pose space. To solve the problem, we follow Perspective-n-Point (PnP) and predict the correspondences between 3D points in canonical space and 2D facial pixels on the input image to solve the 6DoF pose parameters. In this framework, the central problem of 6DoF estimation is building the correspondence matrix between a set of sampled 2D pixels and 3D points, and we propose a Correspondence Learning Transformer (CLT) to achieve this goal. Specifically, we build the 2D and 3D features with local, global, and semantic information, and employ self-attention to make the 2D and 3D features interact with each other and build the 2D–3D correspondence. Besides, we argue that 6DoF estimation is not only related with face appearance itself but also the facial external context, which contains rich information about the distance to camera. Therefore, we extract global-and-local features from the integration of face and context, where the cropped face image with smaller receptive fields concentrates on the small distortion by perspective projection, and the whole image with large receptive field provides shoulder and environment information. Experiments show that our method achieves a 2.0% improvement of$MAE_{r}$and$ADD$on ARKitFace and a 4.0%/0.7% improvement of$MAE_{t}$on ARKitFace/BIWI.

Unsupervised domain adaptation (UDA) is an emerging learning paradigm that models on unlabeled datasets by leveraging model knowledge built on other labeled datasets, in which the statistical distributions of these datasets are usually not identical. Formally, UDA is to leverage knowledge from a labeled source domain to promote an unlabeled target domain. Although there have been a variety of methods proposed to address the UDA problem, most of them are dedicated to single-source-to-single-target domain, while the works on single-source-to-multitarget domain are relatively rare. Compared to the single-source domain with single-target domain scenario, the UDA from single-source domain to multitarget domain is more challenging since it needs to consider not only the relationships between the source and the target domains but also those among the target domains. To this end, this article proposes a kind of dictionary learning-based unsupervised multitarget domain adaptation method (DL-UMTDA). In DL-UMTDA, a common dictionary is constructed to correlate the single-source and multitarget domains, while individual dictionaries are designed to exploit the private knowledge for the target domains. Through learning the corresponding dictionary representation coefficients in the UDA process, the correlations from the source to the target domains as well as these potential relationships between the target domains can be effectively exploited. In addition, we design an alternating algorithm to solve the DL-UMTDA model with theoretical convergence guarantee. Finally, extensive experiments on benchmark (Office + Caltech) and real datasets (AgeDB, Morph, and CACD) validate the superiority of the proposed method.

Exemplar rehearsal-based methods with knowledge distillation (KD) have been widely used in class incremental learning (CIL) scenarios. However, they still suffer from performance degradation because of severely distribution discrepancy between training and test set caused by the limited storage memory on previous classes. In this paper, we mathematically model the data distribution and the discrepancy at the incremental stages with mixture uniform distribution (MUD). Then, we propose the asymmetric mix distillation method to uniformly minimize the error of each class from distribution discrepancy perspective. Specifically, we firstly promote mixup in CIL scenarios with the incremental mix samplers and incremental mix factor to calibrate the raw training data distribution. Next, mix distillation label augmentation is incorporated into the data distribution to inherit the knowledge information from the previous models. Based on the above augmented data distribution, our trained model effectively alleviates the performance degradation and extensive experimental results validate that our method exhibits superior performance on CIL benchmarks.

Compared with the feature-based distillation methods, logits distillation can liberalize the requirements of consistent feature dimension between teacher and student networks, while the performance is deemed inferior in face recognition. One major challenge is that the light-weight student network has difficulty fitting the target logits due to its low model capacity, which is attributed to the significant number of identities in face recognition. Therefore, we seek to probe the target logits to extract the primary knowledge related to face identity, and discard the others, to make the distillation more achievable for the student network. Specifically, there is a tail group with near-zero values in the prediction, containing minor knowledge for distillation. To provide a clear perspective of its impact, we first partition the logits into two groups, i.e., Primary Group and Secondary Group, according to the cumulative probability of the softened prediction. Then, we reorganize the Knowledge Distillation (KD) loss of grouped logits into three parts, i.e., Primary-KD, Secondary-KD, and Binary-KD. Primary-KD refers to distilling the primary knowledge from the teacher, Secondary-KD aims to refine minor knowledge but increases the difficulty of distillation, and Binary-KD ensures the consistency of knowledge distribution between teacher and student. We experimentally found that (1) Primary-KD and Binary-KD are indispensable for KD, and (2) Secondary-KD is the culprit restricting KD at the bottleneck. Therefore, we propose a Grouped Knowledge Distillation (GKD) that retains the Primary-KD and Binary-KD but omits Secondary-KD in the ultimate KD loss calculation. Extensive experimental results on popular face recognition benchmarks demonstrate the superiority of proposed GKD over state-of-the-art methods.

This paper presents a framework for efficient 3D clothed avatar reconstruction. By combining the advantages of the high accuracy of optimization-based methods and the efficiency of learning-based methods, we propose a coarse-to-fine way to realize a high-fidelity clothed avatar reconstruction (CAR) from a single image. At the first stage, we use an implicit model to learn the general shape in the canonical space of a person in a learning-based way, and at the second stage, we refine the surface detail by estimating the non-rigid deformation in the posed space in an optimization way. A hyper-network is utilized to generate a good initialization so that the convergence of the optimization process is greatly accelerated. Extensive experiments on various datasets show that the proposed CAR successfully produces high-fidelity avatars for arbitrarily clothed humans in real scenes. The codes will be released in https://github.com/TingtingLiao/CAR.

Controllability, generalizability and efficiency are the major objectives of constructing face avatars represented by neural implicit field. However, existing methods have not managed to accommodate the three requirements simultaneously. They either focus on static portraits, restricting the representation ability to a specific subject, or suffer from substantial computational cost, limiting their flexibility. In this paper, we propose One-shot Talking face Avatar (OTAvatar), which constructs face avatars by a generalized controllable tri-plane rendering solution so that each personalized avatar can be constructed from only one portrait as the reference. Specifically, OTAvatar first inverts a portrait image to a motion-free identity code. Second, the identity code and a motion code are utilized to modulate an efficient CNN to generate a tri-plane formulated volume, which encodes the subject in the desired motion. Finally, volume rendering is employed to generate an image in any view. The core of our solution is a novel decoupling-by-inverting strategy that disentangles identity and motion in the latent code via optimization-based inversion. Benefiting from the efficient tri-plane representation, we achieve controllable rendering of generalized face avatar at 35 FPS on AIOO. Experiments show promising performance of crossidentity reenactment on subjects out of the training set and better 3D consistency. The code is available at https://github.com/theEricMaIOTAvatar.

The ability to discover abstract physical concepts and understand how they work in the world through observing lies at the core of human intelligence. The acquisition of this ability is based on compositionally perceiving the environment in terms of objects and relations in an unsupervised manner. Recent approaches learn object-centric represen-tations and capture visually observable concepts of objects, e.g., shape, size, and location. In this paper, we take a step forward and try to discover and represent intrinsic physical concepts such as mass and charge. We introduce the PHYsi-cal Concepts Inference NEtwork (PHYCINE), a system that infers physical concepts in different abstract levels with-out supervision. The key insights underlining PHYCINE are two-fold, commonsense knowledge emerges with pre-diction, and physical concepts of different abstract levels should be reasoned in a bottom-up fashion. Empirical eval-uation demonstrates that variables inferred by our system work in accordance with the properties of the corresponding physical concepts. We also show that object representations containing the discovered physical concepts variables could help achieve better performance in causal reasoning tasks, i.e., ComPhy.

The goal of domain generalization (DG) is to enhance the generalization capability of the model learned from a source domain to other unseen domains. The recently developed Sharpness-Aware Minimization (SAM) method aims to achieve this goal by minimizing the sharpness measure of the loss landscape. Though SAM and its variants have demonstrated impressive DG performance, they may not always converge to the desired flat region with a small loss value. In this paper, we present two conditions to ensure that the model could converge to a flat minimum with a small loss, and present an algorithm, named Sharpness-Aware Gradient Matching (SAGM), to meet the two conditions for improving model generalization capability. Specifically, the optimization objective of SAGM will simultaneously minimize the empirical risk, the perturbed loss (i.e., the maximum loss within a neighborhood in the parameter space), and the gap between them. By implicitly aligning the gradient directions between the empirical risk and the perturbed loss, SAGM improves the generalization capability over SAM and its variants without increasing the computational cost. Extensive experimental results show that our proposed SAGM method consistently outperforms the state-of-the-art methods on five DG benchmarks, including PACS, VLCS, OfficeHome, TerraIncognita, and DomainNet. Codes are available at https://github.com/Wang-pengfei/SAGM.

The function of constructing the hierarchy of objects is important to the visual process of the human brain. Previous studies have successfully adopted capsule networks to decompose the digits and faces into parts in an unsupervised manner to investigate the similar perception mechanism of neural networks. However, their descriptions are restricted to the 2D space, limiting their capacities to imitate the intrinsic 3D perception ability of humans. In this paper, we propose an Inverse Graphics Capsule Network (IGC-Net) to learn the hierarchical 3D face representations from large-scale unlabeled images. The core of IGC-Net is a new type of capsule, named graphics capsule, which represents 3D primitives with interpretable parameters in computer graphics (CG), including depth, albedo, and 3D pose. Specifically, IGC-Net first decomposes the objects into a set of semantic-consistent part-level descriptions and then assembles them into object-level descriptions to build the hierarchy. The learned graphics capsules reveal how the neural networks, oriented at visual perception, understand faces as a hierarchy of 3D models. Besides, the discovered parts can be deployed to the unsupervised face segmentation task to evaluate the semantic consistency of our method. Moreover, the part-level descriptions with explicit physical meanings provide insight into the face analysis that originally runs in a black box, such as the importance of shape and texture for face recognition. Experiments on CelebA, BP4D, and Multi-PIE demonstrate the characteristics of our IGC-Net.

Face anti-spoofing is crucial for ensuring the security and reliability of face recognition systems. Several existing face anti-spoofing methods utilize GAN-like networks to detect presentation attacks by estimating the noise pattern of a spoof image and recovering the corresponding genuine image. But GAN’s limited face appearance space results in the denoised faces cannot cover the full data distribution of genuine faces, thereby undermining the generalization performance of such methods. In this work, we present a pioneering attempt to employ diffusion models to denoise a spoof image and restore the genuine image. The difference between these two images is considered as the spoof noise, which can serve as a discriminative cue for face anti-spoofing. We evaluate our proposed method on several intra-testing and inter-testing protocols, where the experimental results showcase the effectiveness of our method in achieving competitive performance in terms of both accuracy and generalization.

Weakly supervised person search aims to jointly detect and match persons with only bounding box annotations. Existing approaches typically focus on improving the features by exploring the relations of persons. However, scale variation problem is a more severe obstacle and under-studied that a person often owns images with different scales (resolutions). For one thing, small-scale images contain less information of a person, thus affecting the accuracy of the generated pseudo labels. For another, different similarities between cross-scale images of a person increase the difficulty of matching. In this paper, we address it by proposing a novel one-step framework, named Self-similarity driven Scale-invariant Learning (SSL). Scale invariance can be explored based on the self-similarity prior that it shows the same statistical properties of an image at different scales. To this end, we introduce a Multi-scale Exemplar Branch to guide the network in concentrating on the foreground and learning scale-invariant features by hard exemplars mining. To enhance the discriminative power of the learned features, we further introduce a dynamic pseudo label prediction that progressively seeks true labels for training. Experimental results on two standard benchmarks, i.e., PRW and CUHK-SYSU datasets, demonstrate that the proposed method can solve scale variation problem effectively and perform favorably against state-of-the-art methods. Code is available at https://github.com/Wangbenzhi/SSL.git.

Sign Language Translation (SLT) is a challenging task due to its cross-domain nature, involving the translation of visual-gestural language to text. Many previous methods employ an intermediate representation, i.e., gloss sequences, to facilitate SLT, thus transforming it into a two-stage task of sign language recognition (SLR) followed by sign language translation (SLT). However, the scarcity of gloss-annotated sign language data, combined with the information bottleneck in the mid-level gloss representation, has hindered the further development of the SLT task. To address this challenge, we propose a novel Gloss-Free SLT based on Visual-Language Pretraining (GFSLT-VLP), which improves SLT by inheriting language-oriented prior knowledge from pre-trained models, without any gloss annotation assistance. Our approach involves two stages: (i) integrating Contrastive Language-Image Pre-training (CLIP) with masked self-supervised learning to create pre-tasks that bridge the semantic gap between visual and textual representations and restore masked sentences, and (ii) constructing an end-to-end architecture with an encoder-decoder-like structure that inherits the parameters of the pre-trained Visual Encoder and Text Decoder from the first stage. The seamless combination of these novel designs forms a robust sign language representation and significantly improves gloss-free sign language translation. In particular, we have achieved unprecedented improvements in terms of BLEU-4 score on the PHOENIX14T dataset (≥+5) and the CSL-Daily dataset (≥+3) compared to state-of-the-art gloss-free SLT methods. Furthermore, our approach also achieves competitive results on the PHOENIX14T dataset when compared with most of the gloss-based methods1.

Transformers have emerged as the superior choice for face recognition tasks, but their insufficient platform acceleration hinders their application on mobile devices. In contrast, Convolutional Neural Networks (CNNs) capitalize on hardware-compatible acceleration libraries. Consequently, it has become indispensable to preserve the distillation efficacy when transferring knowledge from a Transformer-based teacher model to a CNN-based student model, known as Cross-Architecture Knowledge Distillation (CAKD). Despite its potential, the deployment of CAKD in face recognition encounters two challenges: 1) the teacher and student share disparate spatial information for each pixel, obstructing the alignment of feature space, and 2) the teacher network is not trained in the role of a teacher, lacking proficiency in handling distillation-specific knowledge. To surmount these two constraints, 1) we first introduce a Unified Receptive Fields Mapping module (URFM) that maps pixel features of the teacher and student into local features with unified receptive fields, thereby synchronizing the pixel-wise spatial information of teacher and student. Subsequently, 2) we develop an Adaptable Prompting Teacher network (APT) that integrates prompts into the teacher, enabling it to manage distillation-specific knowledge while preserving the model's discriminative capacity. Extensive experiments on popular face benchmarks and two large-scale verification sets demonstrate the superiority of our method.

Face anti-spoofing (FAS) has lately attracted increasing attention due to its vital role in securing face recognition systems from presentation attacks (PAs). As more and more realistic PAs with novel types spring up, early-stage FAS methods based on handcrafted features become unreliable due to their limited representation capacity. With the emergence of large-scale academic datasets in the recent decade, deep learning based FAS achieves remarkable performance and dominates this area. However, existing reviews in this field mainly focus on the handcrafted features, which are outdated and uninspiring for the progress of FAS community. In this paper, to stimulate future research, we present the first comprehensive review of recent advances in deep learning based FAS. It covers several novel and insightful components: 1) besides supervision with binary label (e.g., '0' for bonafide versus '1' for PAs), we also investigate recent methods with pixel-wise supervision (e.g., pseudo depth map); 2) in addition to traditional intra-dataset evaluation, we collect and analyze the latest methods specially designed for domain generalization and open-set FAS; and 3) besides commercial RGB camera, we summarize the deep learning applications under multi-modal (e.g., depth and infrared) or specialized (e.g., light field and flash) sensors. We conclude this survey by emphasizing current open issues and highlighting potential prospects.

Detecting digital face manipulation has attracted extensive attention due to fake media's potential risks to the public. However, recent advances have been able to reduce the forgery signals to a low magnitude. Decomposition, which reversibly decomposes an image into several constituent elements, is a promising way to highlight the hidden forgery details. In this paper, we investigate a novel 3D decomposition based method that considers a face image as the production of the interaction between 3D geometry and lighting environment. Specifically, we disentangle a face image into four graphics components including 3D shape, lighting, common texture, and identity texture, which are respectively constrained by 3D morphable model, harmonic reflectance illumination, and PCA texture model. Meanwhile, we build a fine-grained morphing network to predict 3D shapes with pixel-level accuracy to reduce the noise in the decomposed elements. Moreover, we propose a composition search strategy that enables an automatic construction of an architecture to mine forgery clues from forgery-relevant components. Extensive experiments validate that the decomposed components highlight forgery artifacts, and the searched architecture extracts discriminative forgery features. Thus, our method achieves the state-of-the-art performance.

3D Morphable Model (3DMM) fitting has widely benefited face analysis due to its strong 3D priori. However, previous reconstructed 3D faces suffer from degraded visual verisimilitude due to the loss of fine-grained geometry, which is attributed to insufficient ground-truth 3D shapes, unreliable training strategies and limited representation power of 3DMM. To alleviate this issue, this paper proposes a complete solution to capture the personalized shape so that the reconstructed shape looks identical to the corresponding person. Specifically, given a 2D image as the input, we virtually render the image in several calibrated views to normalize pose variations while preserving the original image geometry. A many-to-one hourglass network serves as the encode-decoder to fuse multiview features and generate vertex displacements as the fine-grained geometry. Besides, the neural network is trained by directly optimizing the visual effect, where two 3D shapes are compared by measuring the similarity between the multiview images rendered from the shapes. Finally, we propose to generate the ground-truth 3D shapes by registering RGB-D images followed by pose and shape augmentation, providing sufficient data for network training. Experiments on several challenging protocols demonstrate the superior reconstruction accuracy of our proposal on the face shape.

Facial age estimation has attracted considerable attention owing to its great potential in applications. However, it still falls short of reliable age estimation due to the lack of sufficient training data with accurate age labels. Using conventional semi-supervised methods to exploit unlabeled data appears to be a good solution, but it does not yield sufficient performance gains while significantly increasing training time. Therefore, to tackle these problems, we present a Divergence-driven Consistency Training (DCT) method for enhancing both efficiency and performance in this paper. Following the idea of pseudo-labeling and consistency regularization, we assign pseudo labels predicted by the teacher model to unlabeled samples and then train the student model on labeled and unlabeled samples based on consistency regularization. Based on this, we propose two main promotions. The first is the Efficient Sample Selection (ESS) strategy, which is based on the Divergence Score to select effective samples from massive unlabeled images to reduce the training time and improve efficiency. The second is Identity Consistency (IC) regularization as the additional loss function, which introduces a high dependency of aging traits on a person. Moreover, we propose Local Prediction (LP), which is a plug-and-play component, to capture local semantics. Extensive experiments on multiple age benchmark datasets, including CACD, Morph II, MIVIA, and Chalearn LAP 2015, indicate DCT outperforms the state-of-the-art approaches significantly.

The availability of handy multi-modal (i.e., RGB-D) sensors has brought about a surge of face anti-spoofing research. However, the current multi-modal face presentation attack detection (PAD) has two defects: (1) The framework based on multi-modal fusion requires providing modalities consistent with the training input, which seriously limits the deployment scenario. (2) The performance of ConvNet-based model on high fidelity datasets is increasingly limited. In this work, we present a pure transformer-based framework, dubbed the Flexible Modal Vision Transformer (FM-ViT), for face anti-spoofing to flexibly target any single-modal (i.e., RGB) attack scenarios with the help of available multi-modal data. Specifically, FM-ViT retains a specific branch for each modality to capture different modal information and introduces the Cross-Modal Transformer Block (CMTB), which consists of two cascaded attentions named Multi-headed Mutual-Attention (MMA) and Fusion-Attention (MFA) to guide each modal branch to mine potential features from informative patch tokens, and to learn modality-agnostic liveness features by enriching the modal information of own CLS token, respectively. Experiments demonstrate that the single model trained based on FM-ViT can not only flexibly evaluate different modal samples, but also outperforms existing single-modal frameworks by a large margin, and approaches the multi-modal frameworks introduced with smaller FLOPs and model parameters.

Face anti-spoofing (FAS) plays a crucial role in securing face recognition systems. Empirically, given an image, a model with more consistent output on different views (i.e., augmentations) of this image usually performs better. Motivated by this exciting observation, we conjecture that encouraging feature consistency of different views may be a promising way to boost FAS models. In this paper, we explore this way thoroughly by enhancing both Embedding-level and Prediction-level Consistency Regularization (EPCR) in FAS. Specifically, at the embedding level, we design a dense similarity loss to maximize the similarities between all positions of two intermediate feature maps in a self-supervised fashion; while at the prediction level, we optimize the mean square error between the predictions of two views. Notably, our EPCR is free of annotations and can directly integrate into semi-supervised learning schemes. Considering different application scenarios, we further design five diverse semi-supervised protocols to measure semi-supervised FAS techniques. We conduct extensive experiments to show that EPCR can significantly improve the performance of several supervised and semi-supervised tasks on benchmark datasets. The codes and protocols are available athttps://github.com/clks-wzz/EPCR.

Facial age estimation has received a lot of attention for its diverse application scenarios. Most existing studies treat each sample equally and aim to reduce the average estimation error for the entire dataset, which can be summarized as General Age Estimation. However, due to the long-tailed distribution prevalent in the dataset, treating all samples equally will inevitably bias the model toward the head classes (usually the adult with a majority of samples). Driven by this, some works suggest that each class should be treated equally to improve performance in tail classes (with a minority of samples), which can be summarized as Long-tailed Age Estimation. However, Long-tailed Age Estimation usually faces a performance trade-off, i.e., achieving improvement in tail classes by sacrificing the head classes. In this paper, our goal is to design a unified framework to perform well on both tasks, killing two birds with one stone. To this end, we propose a simple, effective, and flexible training paradigm named GLAE, which is two-fold. First, we propose Feature Rearrangement (FR) and Pixel-level Auxiliary learning (PA) for better feature utilization to improve the overall age estimation performance. Second, we propose Adaptive Routing (AR) for selecting the appropriate classifier to improve performance in the tail classes while maintaining the head classes. Moreover, we introduce a new metric, named Class-wise Mean Absolute Error (CMAE), to equally evaluate the performance of all classes. Our GLAE provides a surprising improvement on Morph II, reaching the lowest MAE and CMAE of 1.14 and 1.27 years, respectively. Compared to the previous best method, MAE dropped by up to 34%, which is an unprecedented improvement, and for the first time, MAE is close to 1 year old. Extensive experiments on other age benchmark datasets, including CACD, MIVIA, and Chalearn LAP 2015, also indicate that GLAE outperforms the state-of-the-art approaches significantly.

In 3D face reconstruction, orthogonal projection has been widely employed to substitute perspective projection to simplify the fitting process. This approximation performs well when the distance between camera and face is far enough. However, in some scenarios that the face is very close to camera or moving along the camera axis, the methods suffer from the inaccurate reconstruction and unstable temporal fitting due to the distortion under the perspective projection. In this paper, we aim to address the problem of single-image 3D face reconstruction under perspective projection. Specifically, a deep neural network, Perspective Network (PerspNet), is proposed to simultaneously reconstruct 3D face shape in canonical space and learn the correspondence between 2D pixels and 3D points, by which the 6DoF (6 Degrees of Freedom) face pose can be estimated to represent perspective projection. Besides, we contribute a large ARKitFace dataset to enable the training and evaluation of 3D face reconstruction solutions under the scenarios of perspective projection, which has 902,724 2D facial images with ground-truth 3D face mesh and annotated 6DoF pose parameters. Experimental results show that our approach outperforms current state-of-the-art methods by a significant margin. The code and data are available at https://github.com/cbsropenproject/6dof_face.

KNN has gained popularity in machine learning due to its simplicity and good performance. However, kNN faces two problems with classification tasks. The first is that an appropriate distance measurement is required to compute distances between test sample and training samples. The other is the highly computational complexity due to the requirement of searching the nearest neighbors in the whole training data. In order to mitigate these two problems, we propose a novel method named KCNN to enhance the performance of kNN. KCNN uses convolutional neural networks to learn a suitable distance metric as well as prototype reduction to learn a reduced set of prototypes which can represent the original set. It has several superiorities compared with related methods. The combination of CNN and kNN empowers it to extract discriminative hierarchical features with which kNN can easily classify. KCNN learns spatial information on an image instead of considering it as a vector to learn distance metric. Moreover, KCNN simultaneously learns a reduced set of prototypes, which help improve classification efficiency and avoid noisy samples of the massive training set. The proposed method has a better robustness and convergence than CNN, especially when projecting input data into a low-dimension space.

In this paper, a Part-aware Relation Modeling (PRM) is developed to handle the task of human parsing. For pixel-level recognition, it is essential to generate features with adaptive context for various sizes and shapes of human parts. To address the issue, we adaptively capture contexts based on the part-aware relation mechanism. PRM mainly consists of three modules, including a part class module, a part-relation aggregation module, and a part-relation dispersion module. The part class module selectively enhances spatial details of the high-level features to obtain enhanced original features, and then extracts the high-level representations of every human part from a categorical perspective. The part-relation aggregation module is developed to extract the representative global context by exploring associated semantics of human parts, adaptively augmenting the context for human parts. The part-relation dispersion module is designed to generate the discriminative and effective local context and neglect the distracting one by making the affinity of human parts disperse. It ensures that features of the same class will be close to each other and away from those of different classes. By fusing the outputs of the two part-relation modules and the first outputs of the part class module, our PRM produces adaptive contextual features for diverse sizes of human parts, boosting the parsing accuracy. Extensive experiments are conducted to validate the effectiveness of our network, and a new state-of-the-art segmentation performance is achieved on three challenging human parsing datasets,i.e., PASCAL-Person-Part, LIP, and CIHP. PRM is also extended to other tasks like animal parsing, and exhibits its generality.

Discovering the underneath causal relations is the fundamental ability for reasoning about the surrounding environment and predicting the future states in the physical world. Counterfactual prediction from visual input, which requires simulating future states based on unrealized situations in the past, is a vital component in causal relation tasks. In this paper, we work on the confounders that have effect on the physical dynamics, including masses, friction coefficients, etc., to bridge relations between the intervened variable and the affected variable whose future state may be altered. We propose a neural network framework combining Global Causal Relation Attention (GCRA) and Confounder Transmission Structure (CTS). The GCRA looks for the latent causal relations between different variables and estimates the confounders by capturing both spatial and temporal information. The CTS integrates and transmits the learnt confounders in a residual way, so that the estimated confounders can be encoded into the network as a constraint for object positions when performing counterfactual prediction. Without any access to ground truth information about confounders, our model outperforms the state-of-the-art method on various benchmarks by fully utilizing the constraints of confounders. Extensive experiments demonstrate that our model can generalize to unseen environments and maintain good performance.

The networks trained on the long-tailed dataset vary remarkably, despite the same training settings, which shows the great uncertainty in long-tailed learning. To alleviate the uncertainty, we propose a Nested Collaborative Learning (NCL), which tackles the problem by collaboratively learning multiple experts together. NCL consists of two core components, namely Nested Individual Learning (NIL) and Nested Balanced Online Distillation (NBOD), which focus on the individual supervised learning for each single expert and the knowledge transferring among multiple experts, respectively. To learn representations more thoroughly, both NIL and NBOD are formulated in a nested way, in which the learning is conducted on not just all categories from a full perspective but some hard categories from a partial perspective. Regarding the learning in the partial perspective, we specifically select the negative categories with high predicted scores as the hard categories by using a proposed Hard Category Mining (HCM). In the NCL, the learning from two perspectives is nested, highly related and complementary, and helps the network to capture not only global and robust features but also meticulous distinguishing ability. Moreover, self-supervision is further utilized for feature enhancement. Extensive experiments manifest the superiority of our method with outperforming the state-of-the-art whether by using a single model or an ensemble. Code is available at https://github.com/Bazinga699/NCL

Capsule networks are designed to present the objects by a set of parts and their relationships, which provide an insight into the procedure of visual perception. Although recent works have shown the success of capsule networks on simple objects like digits, the human faces with homologous structures, which are suitable for capsules to describe, have not been explored. In this paper, we propose a Hierarchical Parsing Capsule Network (HP-Capsule) for unsupervised face subpart-part discovery. When browsing large-scale face images without labels, the network first encodes the frequently observed patterns with a set of explainable subpart capsules. Then, the subpart capsules are assembled into part-level capsules through a Transformer-based Parsing Module (TPM) to learn the compositional relations between them. During training as the face hierarchy is progressively built and refined, the part capsules adaptively encode the face parts with semantic consistency. HP-Capsule extends the application of capsule networks from digits to human faces and takes a step forward to show how the neural networks understand homologous objects without human intervention. Besides, HP-Capsule gives unsupervised face segmentation results by the covered regions of part capsules, enabling qualitative and quantitative evaluation. Experiments on BP4D and Multi-PIE datasets show the effectiveness of our method.

Decoupling spatiotemporal representation refers to decomposing the spatial and temporal features into dimension-independent factors. Although previous RGB-D-based motion recognition methods have achieved promising performance through the tightly coupled multi-modal spatiotemporal representation, they still suffer from (i) optimization difficulty under small data setting due to the tightly spatiotemporal-entangled modeling; (ii) information redundancy as it usually contains lots of marginal information that is weakly relevant to classification; and (iii) low interaction between multi-modal spatiotemporal information caused by insufficient late fusion. To alleviate these drawbacks, we propose to decouple and recouple spatiotemporal representation for RGB-D-based motion recognition. Specifically, we disentangle the task of learning spatiotemporal representation into 3 sub-tasks: (1) Learning high-quality and dimension independent features through a decoupled spatial and temporal modeling network. (2) Recoupling the decoupled representation to establish stronger space-time dependency. (3) Introducing a Cross-modal Adaptive Posterior Fusion (CAPF) mechanism to capture cross-modal spatiotemporal information from RGB-D data. Seamless combination of these novel designs forms a robust spatiotemporal representation and achieves better performance than state-of-the-art methods on four public motion datasets. Our code is available at https://github.com/damo-cv/MotionRGBD.

This paper presents the summary report on our DFGC 2022 competition. The DeepFake is rapidly evolving, and realistic face-swaps are becoming more deceptive and difficult to detect. On the other hand, methods for detecting DeepFakes are also improving. There is a two-party game between DeepFake creators and defenders. This competition provides a common platform for benchmarking the game between the current state-of-the-arts in Deep-Fake creation and detection methods. The main research question to be answered by this competition is the current state of the two adversaries when competed with each other. This is the second edition after the last year's DFGC 2021, with a new, more diverse video dataset, a more realistic game setting, and more reasonable evaluation metrics. With this competition, we aim to stimulate research ideas for building better defenses against the DeepFake threats. We also release our DFGC 2022 dataset contributed by both our participants and ourselves to enrich the DeepFake data resources for the research community (https://github.com/NiCE-X/DFGC-2022).

Abstract Object re‐identification, such as vehicle re‐identification or pedestrian re‐identification, plays a significant role in intelligent video surveillance systems for public security. Due to viewpoint variations and appearance changes, both pedestrians and vehicles usually have complex intra‐class variations. However, most existing object re‐identification methods often use a sample‐level triplet loss function cooperating with a single‐proxy softmax loss function, which could not handle complex intra‐class variations well. In this paper, a sample‐proxy dual triplet (SPDT) loss function is proposed, which works with a multi‐proxy softmax (MPS) loss function. The MPS loss function is in charge of learning multiple proxies to represent a class. The SPDT loss function is responsible for enlarging inter‐class distances as well as shrinking intra‐class distances on both sample and proxy levels. Therefore, the method not only handles multi‐proxy intra‐class variations but also fully learns discrimination on samples and proxies. Experiments on two large datasets, that is, VeRi776 and DukeMTMC‐reID, demonstrate that the method is superior to state‐of‐the‐art object re‐identification approaches.

In general, vehicle images have varying resolutions due to vehicles’ movements and different camera settings. However, most existing vehicle reidentification models are single-resolution deep networks trained with preuniformly resizing vehicle images, which underestimate adverse effects of varying resolutions and lead to unsatisfactory performance. A straightforward solution for dealing with varying resolutions is to train multiple vehicle reidentification models. Each model is independently trained with images of a specific resolution. However, this straightforward solution requires significant overhead and ignores intrinsic associations among different resolution images. For that, an efficient multiresolution network (EMRN) is proposed for vehicle reidentification in this article. First, EMRN embeds a newly designed multiresolution feature dimension uniform module (MR-FDUM) behind a traditional backbone network (i.e., ResNet-50). As a result, the whole model can extract fixed dimensional features from different resolution images so that it can be trained with one loss function of fixed dimensional parameters rather than training multiple models. Second, a multiresolution image randomly feeding strategy is designed to train EMRN, making each minibatch data of a random resolution during the training process. Consequently, EMRN can implicitly learn collaborative multiresolution features via only a unitary deep network. The experiments on three large-scale data sets, i.e., VeRi776, VehicleID, and VRIC, demonstrate that EMRN is superior to state-of-the-art vehicle reidentification methods.

Face anti-spoofing (FAS) secures face recognition from presentation attacks (PAs). Existing FAS methods usually supervise PA detectors with handcrafted binary or pixel-wise labels. However, handcrafted labels may are not the most adequate way to supervise PA detectors learning sufficient and intrinsic spoofing cues. Instead of using the handcrafted labels, we propose a novel Meta-Teacher FAS (MT-FAS) method to train a meta-teacher for supervising PA detectors more effectively. The meta-teacher is trained in a bi-level optimization manner to learn the ability to supervise the PA detectors learning rich spoofing cues. The bi-level optimization contains two key components: 1) a lower-level training in which the meta-teacher supervises the detector's learning process on the training set; and 2) a higher-level training in which the meta-teacher's teaching performance is optimized by minimizing the detector's validation loss. Our meta-teacher differs significantly from existing teacher-student models because the meta-teacher is explicitly trained for better teaching the detector (student), whereas existing teachers are trained for outstanding accuracy neglecting teaching ability. Extensive experiments on five FAS benchmarks show that with the proposed MT-FAS, the trained meta-teacher 1) provides better-suited supervision than both handcrafted labels and existing teacher-student models; and 2) significantly improves the performances of PA detectors.

The COVID-19 pandemic makes wearing masks mandatory in supermarkets, pharmacies, public transport, etc. Existing facial recognition systems encounter severe performance degradation as the masks occlude key facial regions. Recently, simulation-based methods are proposed to generate masked faces from unmasked faces. However, among simulated faces, there are low-quality samples with negative occlusion, which leads to ambiguous or absent facial features. In this paper, we propose a consistent sub-decision network to obtain sub-decisions that correspond to different facial regions and constrain sub-decisions by weighted bidirectional KL divergence to make the network concentrate on the upper faces without occlusion. In addition, we perform knowledge distillation to drive the masked face embeddings towards an approximation of the original data distribution to mitigate the information loss. Experiments show that the proposed method performs better than the baseline on public masked face recognition datasets, i.e., RMFD, MFR2, and MLFW.

Face presentation attack detection (PAD) is essential to secure face recognition systems primarily from high-fidelity mask attacks. Most existing 3D mask PAD benchmarks suffer from several drawbacks: 1) a limited number of mask identities, types of sensors, and a total number of videos; 2) low-fidelity quality of facial masks. Basic deep models and remote photoplethysmography (rPPG) methods achieved acceptable performance on these benchmarks but still far from the needs of practical scenarios. To bridge the gap to real-world applications, we introduce a large-scale High-Fidelity Mask dataset, namely HiFiMask. Specifically, a total amount of 54,600 videos are recorded from 75 subjects with 225 realistic masks by 7 new kinds of sensors. Along with the dataset, we propose a novel Contrastive Context-aware Learning (CCL) framework. CCL is a new training methodology for supervised PAD tasks, which is able to learn by leveraging rich contexts accurately (e.g., subjects, mask material and lighting) among pairs of live faces and high-fidelity mask attacks. Extensive experimental evaluations on HiFiMask and three additional 3D mask datasets demonstrate the effectiveness of our method. The codes and dataset will be released soon.

In our daily life, a large number of activities require identity verification, e.g., ePassport gates. Most of those verification systems recognize who you are by matching the ID document photo (ID face) to your live face image (spot face). The ID vs. Spot (IvS) face recognition is different from general face recognition where each dataset usually contains a small number of subjects and sufficient images for each subject. In IvS face recognition, the datasets usually contain massive class numbers (million or more) while each class only has two image samples (one ID face and one spot face), which makes it very challenging to train an effective model (e.g., excessive demand on GPU memory if conducting the classification on such massive classes, hardly capture the effective features for bisample data of each identity, etc.). To avoid the excessive demand on GPU memory, a two-stage training method is developed, where we first train the model on the dataset in general face recognition (e.g., MS-Celeb-1M) and then employ the metric learning losses (e.g., triplet and quadruplet losses) to learn the features on IvS data with million classes. To extract more effective features for IvS face recognition, we propose two novel algorithms to enhance the network by selecting harder samples for training. Firstly, a Cross-Batch Hard Example Mining (CB-HEM) is proposed to select the hard triplets from not only the current mini-batch but also past dozens of mini-batches (for convenience, we use batch to denote a mini-batch in the following), which can significantly expand the space of sample selection. Secondly, a Pseudo Large Batch (PLB) is proposed to virtually increase the batch size with a fixed GPU memory. The proposed PLB and CB-HEM can be employed simultaneously to train the network, which dramatically expands the selecting space by hundreds of times, where the very hard sample pairs especially the hard negative pairs can be selected for training to enhance the discriminative capability. Extensive comparative evaluations conducted on multiple IvS benchmarks demonstrate the effectiveness of the proposed method.

Face recognition has witnessed significant progress with the advances of deep convolutional neural networks (CNNs), and the central task of which is how to improve the feature discrimination. To this end, several margin-based (e.g., angular, additive and additive angular margins) softmax loss functions have been proposed to increase the feature margin between different classes. However, despite great achievements have been made, they mainly suffer from four issues: 1) They are based on the assumption of well-cleaned training sets, without considering the consequence of noisy labels inherently existing in most of face recognition datasets; 2) They ignore the importance of informative (e.g., semi-hard) features mining for discriminative learning; 3) They encourage the feature margin only from the perspective of ground truth class, without realizing the discriminability from other non-ground truth classes; and 4) They set the feature margin between different classes to be same and fixed, which may not adapt the situation of unbalanced data in different classes very well. To cope with these issues, this paper develops a novel loss function, which explicitly estimates the noisy labels to drop them and adaptively emphasizes the semi-hard feature vectors from the remaining reliable ones to guide the discriminative feature learning. Thus we can address all the above issues and achieve more discriminative features for face recognition. To the best of our knowledge, this is the first attempt to inherit the advantages of feature-based noisy labels detection, feature mining and feature margin into a unified loss function. Extensive experimental results on a variety of face recognition benchmarks have demonstrated the effectiveness of our method over state-of-the-art alternatives. Our source code is available at http://www.cbsr.ia.ac.cn/users/xiaobowang/.

Most existing methods of human parsing still face a challenge: how to extract the accurate foreground from similar or cluttered scenes effectively. In this paper, we propose a Grammar-induced Wavelet Network (GWNet), to deal with the challenge. GWNet mainly consists of two modules, including a blended grammar-induced module and a wavelet prediction module. We design the blended grammar-induced module to exploit the relationship of different human parts and the inherent hierarchical structure of a human body by means of grammar rules in both cascaded and paralleled manner. In this way, conspicuous parts, which are easily distinguished from the background, can amend the segmentation of inconspicuous ones, improving the foreground extraction. We also design a Part-aware Convolutional Recurrent Neural Network (PCRNN) to pass messages which are generated by grammar rules. To further improve the performance, we propose a wavelet prediction module to capture the basic structure and the edge details of a person by decomposing the low-frequency and high-frequency components of features. The low-frequency component can represent the smooth structures and the high-frequency components can describe the fine details. We conduct extensive experiments to evaluate GWNet on PASCAL-Person-Part, LIP, and PPSS datasets. GWNet obtains state-of-the-art performance on these human parsing datasets.

A standard pipeline of current face recognition frameworks consists of four individual steps: locating a face with a rough bounding box and several fiducial landmarks, aligning the face image using a pre-defined template, extracting representations and comparing. Among them, face detection, landmark detection and representation learning have long been studied and a lot of works have been proposed. As an important step with a big impact on recognition performance, the alignment step has attracted little attention. In this paper, we first explore and highlight the effects of different alignment templates on face recognition. Then, for the first time, we try to automatically search for the optimal template. We construct a well-defined searching space by decomposing the template searching into the crop size and vertical shift, and propose an efficient method Face Alignment Policy Search (FAPS). Besides, a well-designed benchmark is proposed to evaluate the searched policy. Experiments on our proposed benchmark validate the effectiveness of our method to improve the face recognition performance.

Detecting digital face manipulation has attracted extensive attention due to fake media’s potential harms to the public. However, recent advances have been able to reduce the forgery signals to a low magnitude. Decomposition, which reversibly decomposes an image into several constituent elements, is a promising way to highlight the hidden forgery details. In this paper, we consider a face image as the production of the intervention of the underlying 3D geometry and the lighting environment, and decompose it in a computer graphics view. Specifically, by disentangling the face image into 3D shape, common texture, identity texture, ambient light, and direct light, we find the devil lies in the direct light and the identity texture. Based on this observation, we propose to utilize facial detail, which is the combination of direct light and identity texture, as the clue to detect the subtle forgery patterns. Besides, we highlight the manipulated region with a supervised attention mechanism and introduce a two-stream structure to exploit both face image and facial detail together as a multi-modality task. Extensive experiments indicate the effectiveness of the extra features extracted from the facial detail, and our method achieves the state-of-the-art performance.

Although the anchor-based detectors have taken a big step forward in pedestrian detection, the overall performance of algorithm still needs further improvement for practical applications, e.g., a good trade-off between the accuracy and efficiency. To this end, this paper proposes a series of systematic optimization strategies for the detection pipeline of one-stage detector, forming a single shot anchor-based detector (SADet) for efficient and accurate pedestrian detection, which includes three main improvements. Firstly, we optimize the sample generation process by assigning soft labels to the outlier samples to generate semi-positive samples with continuous tag value between 0 and 1. Secondly, a novel Center-IoU loss is applied as a new regression loss for bounding box regression, which not only retains the good characteristics of IoU loss, but also solves some defects of it. Thirdly, we also design Cosine-NMS for the post-processing of predicted bounding boxes, and further propose adaptive anchor matching to enable the model to adaptively match the anchor boxes to full or visible bounding boxes according to the degree of occlusion. Though structurally simple, it presents state-of-the-art result and real-time speed of 20 FPS for VGA-resolution images (640×480) tested on one GeForce GTX 1080Ti GPU on challenging pedestrian detection benchmarks, i.e., CityPersons, Caltech, and human detection benchmark CrowdHuman, leading to a new attractive pedestrian detector.

3D human pose and shape recovery from a monocular RGB image is a challenging task. Existing learning based methods highly depend on weak supervision signals, e.g. 2D and 3D joint location, due to the lack of in-the-wild paired 3D supervision. However, considering the 2D-to-3D ambiguities existed in these weak supervision labels, the network is easy to get stuck in local optima when trained with such labels. In this paper, we reduce the ambituity by optimizing multiple initializations. Specifically, we propose a three-stage framework named Multi-Initialization Optimization Network (MION). In the first stage, we strategically select different coarse 3D reconstruction candidates which are compatible with the 2D keypoints of input sample. Each coarse reconstruction can be regarded as an initialization leads to one optimization branch. In the second stage, we design a mesh refinement transformer (MRT) to respectively refine each coarse reconstruction result via a self-attention mechanism. Finally, a Consistency Estimation Network (CEN) is proposed to find the best result from mutiple candidates by evaluating if the visual evidence in RGB image matches a given 3D reconstruction. Experiments demonstrate that our Multi-Initialization Optimization Network outperforms existing 3D mesh based methods on multiple public benchmarks.

For face presentation attack detection (PAD), most of the spoofing cues are subtle, local image patterns (e.g., local image distortion, 3D mask edge and cut photo edges). The representations of existing PAD works with simple global pooling method, however, lose the local feature discriminability. In this paper, the VLAD aggregation method is adopted to quantize local features with visual vocabulary locally partitioning the feature space, and hence preserve the local discriminability. We further propose the vocabulary separation and adaptation method to modify VLAD for cross-domain PAD task. The proposed vocabulary separation method divides vocabulary into domain-shared and domain-specific visual words to cope with the diversity of live and attack faces under the cross-domain scenario.The proposed vocabulary adaptation method imitates the maximization step of the k-means algorithm in the end-to-end training, which guarantees the visual words be close to the center of assigned local features and thus brings robust similarity measurement. We give illustrations and extensive experiments to demonstrate the effectiveness of VLAD with the proposed vocabulary separation and adaptation method on standard cross-domain PAD benchmarks. The codes are available at https://github.com/Liubinggunzu/VLAD-VSA.

Face detection has achieved significant progress in recent years. However, high performance face detection still remains a very challenging problem, especially when there exists many tiny faces. In this paper, we present a single-shot refinement face detector namely RefineFace to achieve high performance. Specifically, it consists of five modules: selective two-step regression (STR), selective two-step classification (STC), scale-aware margin loss (SML), feature supervision module (FSM) and receptive field enhancement (RFE). To enhance the regression ability for high location accuracy, STR coarsely adjusts locations and sizes of anchors from high level detection layers to provide better initialization for subsequent regressor. To improve the classification ability for high recall efficiency, STC first filters out most simple negatives from low level detection layers to reduce search space for subsequent classifier, then SML is applied to better distinguish faces from background at various scales and FSM is introduced to let the backbone learn more discriminative features for classification. Besides, RFE is presented to provide more diverse receptive field to better capture faces in some extreme poses. Extensive experiments conducted on WIDER FACE, AFW, PASCAL Face, FDDB, MAFA demonstrate that our method achieves state-of-the-art results and runs at 37.3 FPS with ResNet-18 for VGA-resolution images.

Although face recognition has made dramatic improvements in recent years, there are still many challenges in real-world applications such as face recognition for the elderly and children, for the surveillance scenes and for Near infrared vs. Visible light (NIR-VIS) heterogeneous scene, etc. Due to the existence of these challenges, there are usually domain gaps between training (source domain) and test (target domain). A common way to improve the performance on the target domain is fine-tuning the base model trained on source domain using target data. However, it will severely degrade performance on the source domain. Another way which jointly trains models using both source and target data, suffers from the heavy computations and large data storage, especially when we continue to encounter new domains. In response to these problems, we introduce a new challenging task: Single Exemplar Domain Incremental Learning (SE-DIL), which utilizes the target domain data and just one exemplar per identity from source domain data to quickly improve the performance on the target domain while keeping the performance on the source domain. To deal with SE-DIL, we propose our Fast Adapting without Forgetting (FAwF) method with three components: margin-based exemplar selection, prototype-based class extension and hard&soft knowledge distillation. Through FAwF, we can well maintain the source domain performance with only one sample per source domain class, greatly reducing the fine-tuning time-cost and data storage. Besides, we collected a large-scale children face dataset KidsFace with 12 K identities for studying the SE-DIL in face recognition. Extensive analysis and experiments on our KidsFace-Test protocol and other challenging face test sets show that our method performs better than the state-of-the-art methods on both target and source domain.

Convolutional neural network based methods have dominated object detection in recent years, which can be divided into the one-stage approach and the two-stage approach. In general, the two-stage approach (e.g., Faster R-CNN) achieves high accuracy, while the one-stage approach (e.g., SSD) has the advantage of high efficiency. To inherit the merits of both while overcoming their disadvantages, we propose a novel single-shot based detector, namely RefineDet++, which achieves better accuracy than two-stage methods and maintains comparable efficiency of one-stage methods. The proposed RefineDet++ consists of two inter-connected modules: the anchor refinement module and the alignment detection module. Specifically, the former module aims to (1) filter out negative anchors to reduce search space for the subsequent classifier, and (2) coarsely adjust the locations and sizes of anchors to provide better initialization for the subsequent regressor. The latter module takes (1) the refined anchors as the input from the former module with (2) a newly designed alignment convolution operation to further improve the regression accuracy and predict multi-class label. Meanwhile, we design a transfer connection block to transfer the features in the anchor refinement module to predict locations, sizes and class labels of objects in the object detection module. The multi-task loss function enables us to train the whole network in an end-to-end way. Extensive experiments on PASCAL VOC and MS COCO demonstrate that RefineDet++ achieves state-of-the-art detection accuracy with high efficiency.

Face recognition systems are sometimes deployed to a target domain with limited unlabeled samples available. For instance, a model trained on the large-scale webfaces may be required to adapt to a NIR-VIS scenario via very limited unlabeled faces. This situation poses a great challenge to Unsupervised Domain Adaptation with Limited samples for Face Recognition (UDAL-FR), which is less studied in previous works. In this paper, with deep learning methods, we propose a novel training remedy by decomposing the model into the weight parameters and the BN statistics in the training phase. Based on decomposing, we design a novel framework via meta-learning, calledDecomposed Meta Batch Normalization(DMBN) for fast domain adaptation in face recognition. DMBN trains the network such that domain-invariant information is prone to store in the weight parameters and domain-specific knowledge tends to be represented by the BN statistics. Specifically, DMBN constructs distribution-shifted tasks via domain-aware sampling, on which several meta-gradients are obtained by optimizing discriminative representations across different BNs. Finally, the weight parameters are updated with these meta-gradients for better consistency across different BNs. With the learned weight parameters, the adaptation is very fast since only the BN updating on limited data is needed. We propose two UDAL-FR benchmarks to evaluate the domain-adaptive ability of a model with limited unlabeled samples. Extensive experiments validate the efficacy of our proposed DMBN.

Face Presentation Attack Detection (PAD) approaches based on multi-modal data have been attracted increasingly by the research community. However, they require multi-modal face data consistently involved in both the training and testing phases. It would severely limit the applicability due to the most Face Anti-spoofing (FAS) systems are only equipped with Visible (VIS) imaging devices, i.e., RGB cameras. Therefore, how to use other modality (i.e., Near-Infrared (NIR)) to assist the performance improvement of VIS-based PAD is significant for FAS. In this work, we first discuss the big gap of performances among different modalities even though the same backbone network is applied. Then, we propose a novel Cross-modal Auxiliary (CMA) framework for the VIS-based FAS task. The main trait of CMA is that the performance can be greatly improved with the help of other modality while no other modality is required in the testing stage. The proposed CMA consists of a Modality Translation Network (MT-Net) and a Modality Assistance Network (MA-Net). The former aims to close the visible gap between different modalities via a generative model that maps inputs from one modality (i.e., RGB) to another (i.e., NIR). The latter focuses on how to use the translated modality (i.e., target modality) and RGB modality (i.e., source modality) together to train a discriminative PAD model. Extensive experiments are conducted to demonstrate that the proposed framework can push the state-of-the-art (SOTA) performances on both multi-modal datasets (i.e., CASIA-SURF, CeFA, and WMCA) and RGB-based datasets (i.e., OULU-NPU, and SiW).

Pedestrian detection in crowded scenes is a challenging problem, because occlusion happens frequently among different pedestrians. In this paper, we propose an effective and efficient detection network to hunt pedestrians in crowd scenes. The proposed method, namely PedHunter, introduces strong occlusion handling ability to existing region-based detection networks without bringing extra computations in the inference stage. Specifically, we design a mask-guided module to leverage the head information to enhance the feature representation learning of the backbone network. Moreover, we develop a strict classification criterion by improving the quality of positive samples during training to eliminate common false positives of pedestrian detection in crowded scenes. Besides, we present an occlusion-simulated data augmentation to enrich the pattern and quantity of occlusion samples to improve the occlusion robustness. As a consequent, we achieve state-of-the-art results on three pedestrian detection datasets including CityPersons, Caltech-USA and CrowdHuman. To facilitate further studies on the occluded pedestrian detection in surveillance scenes, we release a new pedestrian dataset, called SUR-PED, with a total of over 162k high-quality manually labeled instances in 10k images. The proposed dataset, source codes and trained models are available at https://github.com/ChiCheng123/PedHunter.

Head and human detection have been rapidly improved with the development of deep convolutional neural networks. However, these two tasks are often studied separately without considering their inherent correlation, leading to that 1) head detection is often trapped in more false positives, and 2) the performance of human detector frequently drops dramatically in crowd scenes. To handle these two issues, we present a novel joint head and human detection network, namely JointDet, which effectively detects head and human body simultaneously. Moreover, we design a head-body relationship discriminating module to perform relational learning between heads and human bodies, and leverage this learned relationship to regain the suppressed human detections and reduce head false positives. To verify the effectiveness of the proposed method, we annotate head bounding boxes of the CityPersons and Caltech-USA datasets, and conduct extensive experiments on the CrowdHuman, CityPersons and Caltech-USA datasets. As a consequence, the proposed JointDet detector achieves state-of-the-art performance on these three benchmarks. To facilitate further studies on the head and human detection problem, all new annotations, source codes and trained models are available at https://github.com/ChiCheng123/JointDet.

Face anti-spoofing is crucial to the security of face recognition systems. Most previous methods formulate face anti-spoofing as a supervised learning problem to detect various predefined presentation attacks, which need large scale training data to cover as many attacks as possible. However, the trained model is easy to overfit several common attacks and is still vulnerable to unseen attacks. To overcome this challenge, the detector should: 1) learn discriminative features that can generalize to unseen spoofing types from predefined presentation attacks; 2) quickly adapt to new spoofing types by learning from both the predefined attacks and a few examples of the new spoofing types. Therefore, we define face anti-spoofing as a zero- and few-shot learning problem. In this paper, we propose a novel Adaptive Inner-update Meta Face Anti-Spoofing (AIM-FAS) method to tackle this problem through meta-learning. Specifically, AIM-FAS trains a meta-learner focusing on the task of detecting unseen spoofing types by learning from predefined living and spoofing faces and a few examples of new attacks. To assess the proposed approach, we propose several benchmarks for zero- and few-shot FAS. Experiments show its superior performances on the presented benchmarks to existing methods in existing zero-shot FAS protocols.

Long-tailed problem has been an important topic in face recognition task. However, existing methods only concentrate on the long-tailed distribution of classes. Differently, we devote to the long-tailed domain distribution problem, which refers to the fact that a small number of domains frequently appear while other domains far less existing. The key challenge of the problem is that domain labels are too complicated (related to race, age, pose, illumination, etc.) and inaccessible in real applications. In this paper, we propose a novel Domain Balancing (DB) mechanism to handle this problem. Specifically, we first propose a Domain Frequency Indicator (DFI) to judge whether a sample is from head domains or tail domains. Secondly, we formulate a light-weighted Residual Balancing Mapping (RBM) block to balance the domain distribution by adjusting the network according to DFI. Finally, we propose a Domain Balancing Margin (DBM) in the loss function to further optimize the feature space of the tail domains to improve generalization. Extensive analysis and experiments on several face recognition benchmarks demonstrate that the proposed method effectively enhances the generalization capacities and achieves superior performance.

Face recognition systems are usually faced with unseen domains in real-world applications and show unsatisfactory performance due to their poor generalization. For example, a well-trained model on webface data cannot deal with the ID vs. Spot task in surveillance scenario. In this paper, we aim to learn a generalized model that can directly handle new unseen domains without any model updating. To this end, we propose a novel face recognition method via meta-learning named Meta Face Recognition (MFR). MFR synthesizes the source/target domain shift with a meta-optimization objective, which requires the model to learn effective representations not only on synthesized source domains but also on synthesized target domains. Specifically, we build domain-shift batches through a domain-level sampling strategy and get back-propagated gradients/meta-gradients on synthesized source/target domains by optimizing multi-domain distributions. The gradients and meta-gradients are further combined to update the model to improve generalization. Besides, we propose two benchmarks for generalized face recognition evaluation. Experiments on our benchmarks validate the generalization of our method compared to several baselines and other state-of-the-arts. The proposed benchmarks and code will be available at https://github.com/cleardusk/MFR.

Face anti-spoofing is critical to the security of face recognition systems. Depth supervised learning has been proven as one of the most effective methods for face anti-spoofing. Despite the great success, most previous works still formulate the problem as a single-frame multi-task one by simply augmenting the loss with depth, while neglecting the detailed fine-grained information and the interplay between facial depths and moving patterns. In contrast, we design a new approach to detect presentation attacks from multiple frames based on two insights: 1) detailed discriminative clues (e.g., spatial gradient magnitude) between living and spoofing face may be discarded through stacked vanilla convolutions, and 2) the dynamics of 3D moving faces provide important clues in detecting the spoofing faces. The proposed method is able to capture discriminative details via Residual Spatial Gradient Block (RSGB) and encode spatio-temporal information from Spatio-Temporal Propagation Module (STPM) efficiently. Moreover, a novel Contrastive Depth Loss is presented for more accurate depth supervision. To assess the efficacy of our method, we also collect a Double-modal Anti-spoofing Dataset (DMAD) which provides actual depth for each sample. The experiments demonstrate that the proposed approach achieves state-of-the-art results on five benchmark datasets including OULU-NPU, SiW, CASIA-MFSD, Replay-Attack, and the new DMAD. Codes will be available at https://github.com/clks-wzz/FAS-SGTD.

Object detection has been dominated by anchor-based detectors for several years. Recently, anchor-free detectors have become popular due to the proposal of FPN and Focal Loss. In this paper, we first point out that the essential difference between anchor-based and anchor-free detection is actually how to define positive and negative training samples, which leads to the performance gap between them. If they adopt the same definition of positive and negative samples during training, there is no obvious difference in the final performance, no matter regressing from a box or a point. This shows that how to select positive and negative training samples is important for current object detectors. Then, we propose an Adaptive Training Sample Selection (ATSS) to automatically select positive and negative samples according to statistical characteristics of object. It significantly improves the performance of anchor-based and anchor-free detectors and bridges the gap between them. Finally, we discuss the necessity of tiling multiple anchors per location on the image to detect objects. Extensive experiments conducted on MS COCO support our aforementioned analysis and conclusions. With the newly introduced ATSS, we improve state-of-the-art detectors by a large margin to 50.7% AP without introducing any overhead. The code is available at https://github.com/sfzhang15/ATSS.

As a typical cross-modal problem, image-text bi-directional retrieval relies heavily on the joint embedding learning and similarity measure for each image-text pair. It remains challenging because prior works seldom explore semantic correspondences between modalities and semantic correlations in a single modality at the same time. In this work, we propose a unified Context-Aware Attention Network (CAAN), which selectively focuses on critical local fragments (regions and words) by aggregating the global context. Specifically, it simultaneously utilizes global inter-modal alignments and intra-modal correlations to discover latent semantic relations. Considering the interactions between images and sentences in the retrieval process, intra-modal correlations are derived from the second-order attention of region-word alignments instead of intuitively comparing the distance between original features. Our method achieves fairly competitive results on two generic image-text retrieval datasets Flickr30K and MS-COCO.

With the development of mobile devices, it is hopeful and pressing to deploy face recognition and face anti-spoofing (FAS) model on cell phone or portable devices. Most of existing face anti-spoofing methods focus on building computational costly detector for better spoofing face detection performance. However, these detectors are unfriendly to be deployed on the mobile device for real-time FAS applications. In this paper, we propose a neural architecture search (NAS) based method called Auto-FAS, intending to discover well-suitable lightweight networks for mobile-level face anti-spoofing. In Auto-FAS, a special search space is designed to restrict the model's size, and pixel-wise binary supervision is used to improve the model's performance. We demonstrate both the effectiveness and efficiency of the proposed approach on three public benchmark datasets, which shows the potential real-time FAS application for mobile devices.

Facial expression recognition is inherently a challenging task, especially for the in-the-wild images with various occlusions and large pose variations, which may lead to the loss of some crucial information. To address it, in this paper, we propose an attentive hybrid architecture (AHA) which learns global, local and integrated features based on different face regions. Compared with one type of feature, our extracted features own complementary information and can reduce the loss of crucial information. Specifically, AHA contains three branches, where all sub-networks in those branches employ the attention mechanism to further localize the interested pixels/regions. Moreover, we propose a two-step fusion strategy based on LSTM to deeply explore the hidden correlations among different face regions. Extensive experiments on four popular expression databases (i.e., CK+, FER-2013, SFEW 2.0, RAF-DB) show the effectiveness of the proposed method.

Recently, many Near infrared-visible (NIR-VIS) heterogeneous face recognition (HFR) methods have been proposed in the community. But it remains a challenging problem because of the sensing gap along with large pose variations. In this paper, we propose an Exemplar Guided Cross-Spectral Face Hallucination (EGCH) to reduce the domain discrepancy through disentangled representation learning. For each modality, EGCH contains a spectral encoder as well as a structure encoder to disentangle spectral and structure representation, respectively. It also contains a traditional generator that reconstructs the input from the above two representations, and a structure generator that predicts the facial parsing map from the structure representation. Besides, mutual information minimization and maximization are conducted to boost disentanglement and make representations adequately expressed. Then the translation is built on structure representations between two modalities. Provided with the transformed NIR structure representation and original VIS spectral representation, EGCH is capable to produce high-fidelity VIS images that preserve the topology structure of the input NIR while transfer the spectral information of an arbitrary VIS exemplar. Extensive experiments demonstrate that the proposed method achieves more promising results both qualitatively and quantitatively than the state-of-the-art NIR-VIS methods.

Although face recognition has achieved fairly remarkable results in recent years, it heavily relies on large- scale labeled data to train the high-capacity deep convolutional neural networks. It is unrealistic to collect larger labeled datasets to further boost the performance, which requires burdensome and expensive annotation efforts. Meanwhile, there exist numerous unlabeled face images. It is challenging but promising to jointly utilize limited labeled and abundant unlabeled data to obtain higher performance gain, which is the target of semi-supervised learning. In this paper, we propose a bottom- up method, Neighborhood-Aware Attention Network (NAAN), for semi-supervised face recognition. It clusters unlabeled face images by collaboratively predicting pairwise relations based on their neighborhood information, where the neighborhood is defined as a k-hop ego network centered in the given sample called "ego". Considering the different importance of neighbors, we employ the graph attention network to learn the ego's representation. We evaluate our model on two face recognition datasets MegaFace and IJB-A, and it yields favorably comparable performance to the fully-supervised results.

Object reidentification with the goal of matching pedestrian or vehicle images captured from different camera viewpoints is of considerable significance to public security. Quadruple directional deep learning features (QD-DLFs) can comprehensively describe object images. However, the correlation among QD-DLFs is an unavoidable problem, since QD-DLFs are learned with quadruple independent directional deep networks (QIDDNs) driven with the same training data, and each network holds the same basic deep feature learning architecture (BDFLA). The correlation among QD-DLFs is harmful to the complementarity of QD-DLFs, restricting the object reidentification performance. For that, we propose joint quadruple decorrelation directional deep networks (JQD3Ns) to reduce the correlation among the learned QD-DLFs. In order to jointly train JQD3Ns, besides the softmax loss functions, a parameter correlation cost function is proposed to indirectly reduce the correlation among QD-DLFs by enlarging the dissimilarity among the parameters of JQD3Ns. Extensive experiments on three publicly available large-scale data sets demonstrate that the proposed JQD3Ns approach is superior to multiple state-of-the-art object reidentification methods.

In recent years, deep learning (DL) has been successfully and widely applied in the person reidentification (Re-ID). However, the DL-based person Re-ID methods face a bottleneck that the scales of most existing person Re-ID databases are not large enough for training very deep models. To address this problem, a body symmetry and part-locality-guided direct nonparametric deep feature enhancement (DNDFE) method is proposed in this article. Based on the observation that the body symmetry and part locality are two important appearance properties inherited in the upright walking persons, the proposed method designs two nonparametric layers, namely, the body symmetry average pooling and local normalization layers, to construct a DNDFE module to well explore the body symmetry and part locality properties. The proposed DNDFE module could be directly embedded between the traditional deep feature learning module and similarity learning module to enhance the DL features so as to improve the person Re-ID performance. The experimental results have shown that the proposed DNDFE method is superior to multiple state-of-the-art person Re-ID methods in terms of accuracy and efficiency.

In real applications, face recognition systems are always faced with non-face inputs and low-quality faces due to the complicated conditions like mis-detections by face detectors. However, in deep learning based methods, these outliers are always ignored during training phase and the models tend to make unreasonable decisions on these images. For example, matching a texture-rich patch to an old-man face overconfidently. We formulate this challenge on the task of out-of-distribution detection (OOD), where a network must determine whether or not an input is outside of the set on which the network can safely perform. In this paper, we propose to detect out-of-distribution samples based on uncertainty prediction and the L2-norm of features, so as to effectively filter out non-face and low-quality faces. We demonstrate that the proposed method can reliably detect out-of-distribution samples and improve the performance of face recognition, without the need of labelled OOD data.

Since the human face preserves the richest information for recognizing individuals, face recognition has been widely investigated and achieved great success in various applications in the past decades. However, face spoofing attacks (e.g., face video replay attack) remain a threat to modern face recognition systems. Though many effective methods have been proposed for anti-spoofing, we find that the performance of many existing methods is degraded by illuminations. It motivates us to develop illumination-invariant methods for anti-spoofing. In this paper, we propose a two-stream convolutional neural network (TSCNN), which works on two complementary spaces: RGB space (original imaging space) and multi-scale retinex (MSR) space (illumination-invariant space). Specifically, the RGB space contains the detailed facial textures, yet it is sensitive to illumination; MSR is invariant to illumination, yet it contains less detailed facial information. In addition, the MSR images can effectively capture the high-frequency information, which is discriminative for face spoofing detection. Images from two spaces are fed to the TSCNN to learn the discriminative features for anti-spoofing. To effectively fuse the features from two sources (RGB and MSR), we propose an attention-based fusion method, which can effectively capture the complementarity of two features. We evaluate the proposed framework on various databases, i.e., CASIA-FASD, REPLAY-ATTACK, and OULU, and achieve very competitive performance. To further verify the generalization capacity of the proposed strategies, we conduct cross-database experiments, and the results show the great effectiveness of our method.

Unsupervised domain adaptation in person re-identification resorts to labeled source data to promote the model training on target domain, facing the dilemmas caused by large domain shift and large camera variations. The non-overlapping labels challenge that the source domain and the target domain have entirely different persons further increases the re-identification difficulty. In this paper, we propose a novel algorithm to narrow such domain gaps. We derive a camera style adaptation framework to learn the style-based mappings between different camera views, from the target domain to the source domain, and then we can transfer the identity-based distribution from the source domain to the target domain on the camera level. Target camera variations can be captured by the style adaptation method, thus, the re-identification model trained on the target domain can learn target camera-invariant features better. It indicates that the style translator approximates an appropriate metric space for improving feature matching. To overcome the non-overlapping labels challenge and guide the person re-identification model to narrow the gap further, an efficient and effective soft-labeling method is proposed to mine the intrinsic local structure of the target domain through building the connection between GAN-translated source domain and the target domain. Experiment results conducted on real benchmark datasets indicate that our method gets state-of-the-art results.

Existing enhancement methods are empirically expected to help the high-level end computer vision task: however, that is observed to not always be the case in practice. We focus on object or face detection in poor visibility enhancements caused by bad weathers (haze, rain) and low light conditions. To provide a more thorough examination and fair comparison, we introduce three benchmark sets collected in real-world hazy, rainy, and low-light conditions, respectively, with annotated objects/faces. We launched the UG2+ challenge Track 2 competition in IEEE CVPR 2019, aiming to evoke a comprehensive discussion and exploration about whether and how low-level vision techniques can benefit the high-level automatic visual recognition in various scenarios. To our best knowledge, this is the first and currently largest effort of its kind. Baseline results by cascading existing enhancement and detection models are reported, indicating the highly challenging nature of our new data as well as the large room for further technical innovations. Thanks to a large participation from the research community, we are able to analyze representative team solutions, striving to better identify the strengths and limitations of existing mindsets as well as the future directions.

In order to resist the adverse effect of viewpoint variations, we design quadruple directional deep learning networks to extract quadruple directional deep learning features (QD-DLF) of vehicle images for improving vehicle re-identification performance. The quadruple directional deep learning networks are of similar overall architecture, including the same basic deep learning architecture but different directional feature pooling layers. Specifically, the same basic deep learning architecture that is a shortly and densely connected convolutional neural network is utilized to extract the basic feature maps of an input square vehicle image in the first stage. Then, the quadruple directional deep learning networks utilize different directional pooling layers, i.e., horizontal average pooling layer, vertical average pooling layer, diagonal average pooling layer, and anti-diagonal average pooling layer, to compress the basic feature maps into horizontal, vertical, diagonal, and anti-diagonal directional feature maps, respectively. Finally, these directional feature maps are spatially normalized and concatenated together as a quadruple directional deep learning feature for vehicle re-identification. The extensive experiments on both VeRi and VehicleID databases show that the proposed QD-DLF approach outperforms multiple state-of-the-art vehicle re-identification methods.

High performance face detection remains a very challenging problem, especially when there exists many tiny faces. This paper presents a novel single-shot face detector, named Selective Refinement Network (SRN), which introduces novel twostep classification and regression operations selectively into an anchor-based face detector to reduce false positives and improve location accuracy simultaneously. In particular, the SRN consists of two modules: the Selective Two-step Classification (STC) module and the Selective Two-step Regression (STR) module. The STC aims to filter out most simple negative anchors from low level detection layers to reduce the search space for the subsequent classifier, while the STR is designed to coarsely adjust the locations and sizes of anchors from high level detection layers to provide better initialization for the subsequent regressor. Moreover, we design a Receptive Field Enhancement (RFE) block to provide more diverse receptive field, which helps to better capture faces in some extreme poses. As a consequence, the proposed SRN detector achieves state-of-the-art performance on all the widely used face detection benchmarks, including AFW, PASCAL face, FDDB, and WIDER FACE datasets. Codes will be released to facilitate further studies on the face detection problem.

Benefiting from publicly available databases, face anti-spoofing has recently gained extensive attention in the academic community. However, most of the existing databases focus on the 2D object attacks, including photo and video attacks. The only two public 3D mask face anti-spoofing database are very small. In this paper, we release a multi-modality 3D mask face anti-spoofing database named 3DMA, which contains 920 videos of 67 genuine subjects wearing 48 kinds of 3D masks, captured in visual (VIS) and near-infrared (NIR) modalities. To simulate the real world scenarios, two illumination and four capturing distance settings are deployed during the collection process. To the best of our knowledge, the proposed database is currently the most extensive public database for 3D mask face anti-spoofing. Furthermore, we build three protocols for performance evaluation under different illumination conditions and distances. Experimental results with Convolutional Neural Network (CNN) and LBP-based methods reveal that our proposed 3DMA is indeed a challenge for face anti-spoofing. This database is available at http://www.cbsr.ia.ac.cn/english/3DMA.html. We hope our public 3DMA database can help to pave the way for further research on 3D mask face anti-spoofing.

Recently, deep learning based facial landmark detection has achieved great success. Despite this, we notice that the semantic ambiguity greatly degrades the detection performance. Specifically, the semantic ambiguity means that some landmarks (e.g. those evenly distributed along the face contour) do not have clear and accurate definition, causing inconsistent annotations by annotators. Accordingly, these inconsistent annotations, which are usually provided by public databases, commonly work as the ground-truth to supervise network training, leading to the degraded accuracy. To our knowledge, little research has investigated this problem. In this paper, we propose a novel probabilistic model which introduces a latent variable, i.e. the `real' ground-truth which is semantically consistent, to optimize. This framework couples two parts (1) training landmark detection CNN and (2) searching the `real' ground-truth. These two parts are alternatively optimized: the searched `real' ground-truth supervises the CNN training; and the trained CNN assists the searching of `real' ground-truth. In addition, to recover the unconfidently predicted landmarks due to occlusion and low quality, we propose a global heatmap correction unit (GHCU) to correct outliers by considering the global face shape as a constraint. Extensive experiments on both image-based (300W and AFLW) and video-based (300-VW) databases demonstrate that our method effectively improves the landmark detection accuracy and achieves the state of the art performance.

Training large-scale unbalanced data is the central topic in face recognition. In the past two years, face recognition has achieved remarkable improvements due to the introduction of margin based Softmax loss. However, these methods have an implicit assumption that all the classes possess sufficient samples to describe its distribution, so that a manually set margin is enough to equally squeeze each intra-class variations. However, real face datasets are highly unbalanced, which means the classes have tremendously different numbers of samples. In this paper, we argue that the margin should be adapted to different classes. We propose the Adaptive Margin Softmax to adjust the margins for different classes adaptively. In addition to the unbalance challenge, face data always consists of large-scale classes and samples. Smartly selecting valuable classes and samples to participate in the training makes the training more effective and efficient. To this end, we also make the sampling process adaptive in two folds: Firstly, we propose the Hard Prototype Mining to adaptively select a small number of hard classes to participate in classification. Secondly, for data sampling, we introduce the Adaptive Data Sampling to find valuable samples for training adaptively. We combine these three parts together as AdaptiveFace. Extensive analysis and experiments on LFW, LFW BLUFR and MegaFace show that our method performs better than state-of-the-art methods using the same network architecture and training dataset. Code is available at https://github.com/haoliu1994/AdaptiveFace.

In this paper, we propose an unsupervised graph association (UGA) framework to learn the underlying view-invariant representations from the video pedestrian tracklets. The core points of UGA are mining the underlying cross-view associations and reducing the damage of noise associations. To this end, UGA is adopts a two-stage training strategy: (1) intra-camera learning stage and (2) intercamera learning stage. The former learns the intra-camera representation for each camera. While the latter builds a cross-view graph (CVG) to associate different cameras. By doing this, we can learn view-invariant representation for all person. Extensive experiments and ablation studies on seven re-id datasets demonstrate the superiority of the proposed UGA over most state-of-the-art unsupervised and domain adaptation re-id methods.

Multispectral pedestrian detection has shown great advantages under poor illumination conditions, since the thermal modality provides complementary information for the color image. However, real multispectral data suffers from the position shift problem, i.e. the color-thermal image pairs are not strictly aligned, making one object has different positions in different modalities. In deep learning based methods, this problem makes it difficult to fuse the feature maps from both modalities and puzzles the CNN training. In this paper, we propose a novel Aligned Region CNN (AR-CNN) to handle the weakly aligned multispectral data in an end-to-end way. Firstly, we design a Region Feature Alignment (RFA) module to capture the position shift and adaptively align the region features of the two modalities. Secondly, we present a new multimodal fusion method, which performs feature re-weighting to select more reliable features and suppress the useless ones. Besides, we propose a novel RoI jitter strategy to improve the robustness to unexpected shift patterns of different devices and system settings. Finally, since our method depends on a new kind of labelling: bounding boxes that match each modality, we manually relabel the KAIST dataset by locating bounding boxes in both modalities and building their relationships, providing a new KAIST-Paired Annotation. Extensive experimental validations on existing datasets are performed, demonstrating the effectiveness and robustness of the proposed method. Code and data are available at https://github.com/luzhang16/AR-CNN.

Face recognition methods have achieved high accuracy when faces are captured in frontal pose and constrained scenes. However, severe drop in accuracy is observed when large pose variations exist. The main reason is that the large yaw angle leads to ID information loss. In this paper, we intend to solve the large pose variations in a generation manner. Specifically, we propose a Pose-Weighted Generative Adversarial Network (PW-GAN) for photorealistic frontal view synthesis. We find frontalizing the faces in large poses (yaw angle larger than 60°) is so difficult that the results are not photorealistic and the ID information is lost. To simplify the problem, we first frontalize the face image through 3D face model, which is then used to guide the network predicting. Second, we refine the pose code in the loss function to make the network pay more attention to large poses. Quantitative and qualitative experimental results on the Multi-PIE and LFW demonstrate our method achieves state of the art.

Person Re-Identification (Re-ID) has witnessed great improvements due to the advances of the deep convolutional neural networks (CNN). Despite this, existing methods mainly suffer from the poor generalization ability to unseen scenes because of the different characteristics between different domains. To address this issue, a Clustering and Dynamic Sampling (CDS) method is proposed in this paper, which tries to transfer the useful knowledge of existing labeled source domain to the unlabeled target one. Specifically, to improve the discriminability of CNN model on source domain, we use the commonly shared pedestrian attributes (e.g., gender, hat and clothing color etc.) to enrich the information and resort to the margin-based softmax (e.g., A-Softmax) loss to train the model. For the unlabeled target domain, we iteratively cluster the samples into several centers and dynamically select informative ones from each center to fine-tune the source-domain model. Extensive experiments on DukeMTMC-reID and Market-1501 datasets show that the proposed method greatly improves the state of the arts in unsupervised domain adaptation.

Face alignment, which fits a face model to an image and extracts the semantic meanings of facial pixels, has been an important topic in the computer vision community. However, most algorithms are designed for faces in small to medium poses (yaw angle is smaller than 45 degree), which lack the ability to align faces in large poses up to 90 degree. The challenges are three-fold. First, the commonly used landmark face model assumes that all the landmarks are visible and is therefore not suitable for large poses. Second, the face appearance varies more drastically across large poses, from the frontal view to the profile view. Third, labelling landmarks in large poses is extremely challenging since the invisible landmarks have to be guessed. In this paper, we propose to tackle these three challenges in an new alignment framework termed 3D Dense Face Alignment (3DDFA), in which a dense 3D Morphable Model (3DMM) is fitted to the image via Cascaded Convolutional Neural Networks. We also utilize 3D information to synthesize face images in profile views to provide abundant samples for training. Experiments on the challenging AFLW database show that the proposed approach achieves significant improvements over the state-of-the-art methods.

Re-ranking algorithms can often boost the performance of close-set person re-identification. However, limited efforts have been devoted to answering whether a similar conclusion could be derived on open-set person re-identification. Considering that open-set scenario is more practical in real applications, in this paper, we try to answer this question and do a benchmark study of re-ranking on open-set person re-identification. Specifically, we evaluate three feature descriptors, namely MB-LBP, LOMO, and IDE, and four distance metrics, namely Euclidean, Cosine, RRDA, and XQDA, with their combinations as baseline algorithms. Then, we evaluate four popular re-ranking algorithms, including k-reciprocal Encoding, ECN-3, ECN-4, and DaF. Through extensive benchmark studies on the OPeRIDv1.0 dataset, the results show that re-ranking algorithms, though useful for closed-set person re-identification, are not generally effective for the open-set person re-identification problem. We argue that this is because re-ranking algorithms change the score distributions per query, and hence disrupt the FAR estimation across all queries. Accordingly, we propose to align the re-ranking scores to the original score via the min-max normalization, which verifies our hypothesis above.

Person re-identification is an important technique towards automatic search of a person's presence in a surveillance video. It is becoming a hot research topic due to its value in both machine learning research and video surveillance applications. Considering the current success of deep learning, having tons of person images with identity labels are important and helpful for learning effective person matchers. However, collecting labeled images for person re-identification is more difficult than other similar tasks such as face recognition due to complex intra-class variations in illumination, pose, viewpoint, blur, low resolution, and occlusion. Although the volume of surveillance videos has become larger and larger today, it is time-consuming and costs lots of human labors in labeling a large dataset for person re-identification. In this paper, we propose a semi-automatic data annotation tool to accelerate annotation of person images across multi cameras. This tool consists of automatic person detection and tracking algorithms for person image collection, and an ad-hoc person matcher for automatic person matching suggestions across multi cameras. Moreover, we further utilize background and video sequence information for identity confirmation during annotation, which is also a good intuition for the future design of person re-identification algorithms.

For object detection, the two-stage approach (e.g., Faster R-CNN) has been achieving the highest accuracy, whereas the one-stage approach (e.g., SSD) has the advantage of high efficiency. To inherit the merits of both while overcoming their disadvantages, in this paper, we propose a novel single-shot based detector, called RefineDet, that achieves better accuracy than two-stage methods and maintains comparable efficiency of one-stage methods. RefineDet consists of two inter-connected modules, namely, the anchor refinement module and the object detection module. Specifically, the former aims to (1) filter out negative anchors to reduce search space for the classifier, and (2) coarsely adjust the locations and sizes of anchors to provide better initialization for the subsequent regressor. The latter module takes the refined anchors as the input from the former to further improve the regression accuracy and predict multi-class label. Meanwhile, we design a transfer connection block to transfer the features in the anchor refinement module to predict locations, sizes and class labels of objects in the object detection module. The multitask loss function enables us to train the whole network in an end-to-end way. Extensive experiments on PASCAL VOC 2007, PASCAL VOC 2012, and MS COCO demonstrate that RefineDet achieves state-of-the-art detection accuracy with high efficiency. Code is available at https://github.com/sfzhang15/RefineDet.

A fast exact nearest neighbor search algorithm over large scale data is proposed based on semi-convex hull tree, where each node represents a semi-convex hull, which is made of a set of hyper planes. When performing the task of nearest neighbor queries, unnecessary distance computations can be greatly reduced by quadratic programming. GPUs are also used to accelerate the query process. Experiments conducted on both Intel(R) HD Graphics 4400 and Nvidia Geforce GTX1050 TI, as well as theoretical analysis show that the proposed algorithm yields significant improvements and outperforms current k-d tree based nearest neighbor query algorithms and others.

Subspace clustering refers to the problem of grouping data into their underlying groups. To address this task, spectral clustering based technique is arguably one of the most popular approaches, and its performance largely depends on the constructed similarity. However, most existing works merely employ the primary representation (e.g., sparse or low-rank representation) as the similarity. In this paper, we propose to explore a high-level co-referenced similarity by employing the Hilbert-Schmidt Independence Criterion (HSIC). Moreover, geometry interpretation of the advantage of our co-referenced similarity is provided. Representation-induced kernels such as Mahalanobis metric, can also be easily embedded into the formulation. Extensive experiments on both synthetic and real-world data are conducted to show the superiority of the proposed method over the state-of-the-art alternatives.

In this paper, we propose a shortly and densely connected convolutional neural network (SDC-CNN) for vehicle re-identification. The proposed SDC-CNN mainly consists of short and dense units (SDUs), necessary pooling and normalization layers. The main contribution lies at the design of short and dense connection mechanism, which would effectively improve the feature learning ability. Specifically, in the proposed short and dense connection mechanism, each SDU contains a short list of densely connected convolutional layers and each convolutional layer is of the same appropriate channels. Consequently, the number of connections and the input channel of each convolutional layer are limited in each SDU, and the architecture of SDC-CNN is simple. Extensive experiments on both VeRi and VehicleID datasets show that the proposed SDC-CNN is obviously superior to multiple state-of-the-art vehicle re-identification methods.

Softmax loss is arguably one of the most popular losses to train CNN models for image classification. However, recent works have exposed its limitation on feature discriminability. This paper casts a new viewpoint on the weakness of softmax loss. On the one hand, the CNN features learned using the softmax loss are often inadequately discriminative. We hence introduce a soft-margin softmax function to explicitly encourage the discrmination between different classes. On the other hand, the learned classifier of softmax loss is weak. We propose to assemble multiple these weak classifiers to a strong one, inspired by the recognition that the diversity among weak classifiers is critical to a good ensemble. To achieve the diversity, we adopt the Hilbert-Schmidt Independence Criterion (HSIC). Considering these two aspects in one framework, we design a novel loss, named as Ensemble Soft-Margin Softmax (EM-Softmax). Extensive experiments on benchmark datasets are conducted to show the superiority of our design over the baseline softmax loss and several state-of-the-art alternatives.

Different ages are closely related especially among the adjacent ages because aging is a slow and extremely non-stationary process with much randomness. To explore the relationship between the real age and its adjacent ages, an age group-n encoding (AGEn) method is proposed in this paper. In our model, adjacent ages are grouped into the same group and each age corresponds to n groups. The ages grouped into the same group would be regarded as an independent class in the training stage. On this basis, the original age estimation problem can be transformed into a series of binary classification sub-problems. And a deep Convolutional Neural Networks (CNN) with multiple classifiers is designed to cope with such sub-problems. Later, a Local Age Decoding (LAD) strategy is further presented to accelerate the prediction process, which locally decodes the estimated age value from ordinal classifiers. Besides, to alleviate the imbalance data learning problem of each classifier, a penalty factor is inserted into the unified objective function to favor the minority class. To compare with state-of-the-art methods, we evaluate the proposed method on FG-NET, MORPH II, CACD and Chalearn LAP 2015 databases and it achieves the best performance.

In this letter, we focus on how to boost the performance of person re-identification by exploring the discriminative information among person pairs. A novel dependence-aware feature coding framework is proposed for this task. Specifically, we employ the Hilbert–Schmidt independence criterion as the discriminative term, which is to explore the dependence between different kinds of person pairs, i.e., the same person pairs should be dependence maximized, while the different ones should be dependence minimized. Theoretical discussion and analysis on the convexity of the proposed constraint, as well as the convergence of our algorithm, are provided. Experimental results on two benchmark datasets have demonstrated the advantages of our method over the state-of-the-art alternatives.

Person re-identification (Re-ID) aims to match person images captured from two non-overlapping cameras. In this paper, a deep hybrid similarity learning (DHSL) method for person Re-ID based on a convolution neural network (CNN) is proposed. In our approach, a light CNN learning feature pair for the input image pair is simultaneously extracted. Then, both the elementwise absolute difference and multiplication of the CNN learning feature pair are calculated. Finally, a hybrid similarity function is designed to measure the similarity between the feature pair, which is realized by learning a group of weight coefficients to project the elementwise absolute difference and multiplication into a similarity score. Consequently, the proposed DHSL method is able to reasonably assign complexities of feature learning and metric learning in a CNN, so that the performance of person Re-ID is improved. Experiments on three challenging person Re-ID databases, QMUL GRID, VIPeR, and CUHK03, illustrate that the proposed DHSL method is superior to multiple state-of-the-art person Re-ID methods.

Owing to the variations including both intrinsic and extrinsic factors, age estimation remains a challenging problem. In this paper, five cascaded structure frameworks are proposed for age estimation based on convolutional neural networks. All frameworks are learned and guided by auxiliary demographic information, since other demographic information (i.e., gender and race) is beneficial for age prediction. Each cascaded structure framework is embodied in a parent network and several subnetworks. For example, one of the applied framework is a gender classifier trained by gender information, and then two subnetworks are trained by the male and female samples, respectively. Furthermore, we use the features extracted from the cascaded structure frameworks with Gaussian process regression that can boost the performance further for age estimation. Experimental results on the MORPH II and CACD datasets have gained superior performances compared to the state-of-the-art methods. The mean absolute error is significantly reduced from 3.63 to 2.93 years under the same test protocol on the MORPH II dataset.

Multi-view subspace clustering aims to partition a set of multi-source data into their underlying groups. To boost the performance of multi-view clustering, numerous subspace learning algorithms have been developed in recent years, but with rare exploitation of the representation complementarity between different views as well as the indicator consistency among the representations, let alone considering them simultaneously. In this paper, we propose a novel multi-view subspace clustering model that attempts to harness the complementary information between different representations by introducing a novel position-aware exclusivity term. Meanwhile, a consistency term is employed to make these complementary representations to further have a common indicator. We formulate the above concerns into a unified optimization framework. Experimental results on several benchmark datasets are conducted to reveal the effectiveness of our algorithm over other state-of-the-arts.

Micro emotion recognition is a very challenging problem because of the subtle appearance variants among different facial expression classes. To deal with the mentioned problem, we proposed a multi-modality convolutional neural networks (CNNs) based on visual and geometrical information in this paper. The visual face image and structured geometry are embedded into a unified network and the recognition accuracy can be benefic from the fused information. The proposed network includes two branches. The first branch is used to extract visual feature from color face images, and another branch is used to extract the geometry feature from 68 facial landmarks. Then, both visual and geometry features are concatenated into a long vector. Finally, the concatenated vector is fed to the hinge loss layer. Compared with the CNN architecture only used face images, our method is more effective and has got better performance. In the final testing phase of Micro Emotion Challenge1, our method has got the first place with the misclassification of 80.212137.

Although tremendous strides have been made in face detection, one of the remaining open challenges is to achieve real-time speed on the CPU as well as maintain high performance, since effective models for face detection tend to be computationally prohibitive.To address this challenge, we propose a novel face detector, named FaceBoxes, with superior performance on both speed and accuracy.Specifically, our method has a lightweight yet powerful network structure that consists of the Rapidly Digested Convolutional Layers (RDCL) and the Multiple Scale Convolutional Layers (MSCL).The RDCL is designed to enable Face-Boxes to achieve real-time speed on the CPU.The MSCL aims at enriching the receptive fields and discretizing anchors over different layers to handle faces of various scales.Besides, we propose a new anchor densification strategy to make different types of anchors have the same density on the image, which significantly improves the recall rate of small faces.As a consequence, the proposed detector runs at 20 FPS on a single CPU core and 125 FPS using a GPU for VGA-resolution images.Moreover, the speed of FaceBoxes is invariant to the number of faces.We comprehensively evaluate this method and present stateof-the-art detection performance on several face detection benchmark datasets, including the AFW, PASCAL face, and FDDB.

This paper presents a real-time face detector, named Single Shot Scale-invariant Face Detector (S3FD), which performs superiorly on various scales of faces with a single deep neural network, especially for small faces. Specifically, we try to solve the common problem that anchor-based detectors deteriorate dramatically as the objects become smaller. We make contributions in the following three aspects: 1) proposing a scale-equitable face detection framework to handle different scales of faces well. We tile anchors on a wide range of layers to ensure that all scales of faces have enough features for detection. Besides, we design anchor scales based on the effective receptive field and a proposed equal proportion interval principle; 2) improving the recall rate of small faces by a scale compensation anchor matching strategy; 3) reducing the false positive rate of small faces via a max-out background label. As a consequence, our method achieves state-of-the-art detection performance on all the common face detection benchmarks, including the AFW, PASCAL face, FDDB and WIDER FACE datasets, and can run at 36 FPS on a Nvidia Titan X (Pascal) for VGA-resolution images.

In many face recognition applications, the modalities of face images between the gallery and probe sets are different, which is known as heterogeneous face recognition. How to reduce the feature gap between images from different modalities is a critical issue to develop a highly accurate face recognition algorithm. Recently, joint Bayesian (JB) has demonstrated superior performance on general face recognition compared to traditional discriminant analysis methods like subspace learning. However, the original JB treats the two input samples equally and does not take into account the modality difference between them and may be suboptimal to address the heterogeneous face recognition problem. In this work, we extend the original JB by modeling the gallery and probe images using two different Gaussian distributions to propose a heterogeneous joint Bayesian (HJB) formulation for cross-modality face recognition. The proposed HJB explicitly models the modality difference of image pairs and, therefore, is able to better discriminate the same/different face pairs accurately. Extensive experiments conducted in the case of visible-near-infrared and ID photo versus spot face recognition problems show the superiority of the HJB over previous methods.

In this paper, we propose a novel similarity measure and then introduce an efficient strategy to learn it by using only similar pairs for person verification. Unlike existing metric learning methods, we consider both the difference and commonness of an image pair to increase its discriminativeness. Under a pairconstrained Gaussian assumption, we show how to obtain the Gaussian priors (i.e., corresponding covariance matrices) of dissimilar pairs from those of similar pairs. The application of a log likelihood ratio makes the learning process simple and fast and thus scalable to large datasets. Additionally, our method is able to handle heterogeneous data well. Results on the challenging datasets of face verification (LFW and Pub-Fig) and person re-identification (VIPeR) show that our algorithm outperforms the state-of-the-art methods.

A variety of encoding methods for bag of word (BoW) model have been proposed to encode the local features in image classification. However, most of them are unsupervised and just employ k-means to form the visual vocabulary, thus reducing the discriminative power of the features. In this paper, we propose a metric embedded discriminative vocabulary learning for high-level person representation with application to person re-identification. A new and effective term is introduced which aims at making the same persons closer while different ones farther in the metric space. With the learned vocabulary, we utilize a linear coding method to encode the image-level features (or holistic image features) for extracting high-level person representation. Different from traditional unsupervised approaches, our method can explore the relationship(same or not) among the persons. Since there is an analytic solution to the linear coding, it is easy to obtain the final high-level features. The experimental results on person re-identification demonstrate the effectiveness of our proposed algorithm.

Object detection is a fundamental problem in image understanding. One popular solution is the R-CNN framework [15] and its fast versions [14, 27]. They decompose the object detection problem into two cascaded easier tasks: 1) generating object proposals from images, 2) classifying proposals into various object categories. Despite that we are handling with two relatively easier tasks, they are not solved perfectly and there's still room for improvement. In this paper, we push the "divide and conquer" solution even further by dividing each task into two sub-tasks. We call the proposed method "CRAFT" (Cascade Regionproposal-network And FasT-rcnn), which tackles each task with a carefully designed network cascade. We show that the cascade structure helps in both tasks: in proposal generation, it provides more compact and better localized object proposals, in object classification, it reduces false positives (mainly between ambiguous categories) by capturing both inter-and intra-category variances. CRAFT achieves consistent and considerable improvement over the state-of the-art on object detection benchmarks like PASCAL VOC 07/12 and ILSVRC.

Face alignment, which fits a face model to an image and extracts the semantic meanings of facial pixels, has been an important topic in CV community. However, most algorithms are designed for faces in small to medium poses (below 45), lacking the ability to align faces in large poses up to 90. The challenges are three-fold: Firstly, the commonly used landmark-based face model assumes that all the landmarks are visible and is therefore not suitable for profile views. Secondly, the face appearance varies more dramatically across large poses, ranging from frontal view to profile view. Thirdly, labelling landmarks in large poses is extremely challenging since the invisible landmarks have to be guessed. In this paper, we propose a solution to the three problems in an new alignment framework, called 3D Dense Face Alignment (3DDFA), in which a dense 3D face model is fitted to the image via convolutional neutral network (CNN). We also propose a method to synthesize large-scale training samples in profile views to solve the third problem of data labelling. Experiments on the challenging AFLW database show that our approach achieves significant improvements over state-of-the-art methods.

As digital technologies advance, video has become ubiquitous and hence a rich source of information. Video analytics (or video content analysis) is an important area of computer vision which is concerned with the process of making sense of video content in order to ultimately understand video. Video analytics appears in different forms, such as activity recognition, motion detection, object detection and recognition, person detection and recognition, event and scenario recognition, anomaly detection and identity recognition and verification. Video analytics can be applied in a wide range of domains including healthcare, retail, transport, smart homes, safety and security. The aim of this Special Issue is to raise the awareness of the importance of video analytics. The specific objectives are: (1) to report the latest developments; (2) to identify major research challenges and; (3) to provide visions of future development. A total of 21 papers were submitted to this Special Issue including invited papers and, following a rigorous peer-review process, a total of 11 papers were accepted for inclusion in this Special Issue. The invited paper “Video Analytics Revisited” by Ayesha Choudhary and Santanu Chaudhury presents a concise yet in-depth survey of video analytics. Research problems are discussed and important current applications of video analytics are reviewed. The paper “Human Action Recognition Using Histogram of Motion Intensity and Direction from Multiple Views” by SungYong Chun et al. presents an approach to human activity recognition from multiple views based on estimation of local motion from multiple camera views. A new motion descriptor, histogram of motion intensity and direction, is proposed to capture local motion characteristics of human activity. Classification is done using a support vector machine. Experimental evaluation has demonstrated superior performance of their approach, outperforming 3D optical flow-based approaches with lower computational requirements. The paper “Video Anomaly Detection Using Deep Incremental Slow Feature Analysis Network” by Xing Hu et al. presents an approach to anomaly detection using automatically learned features instead of hand-crafted features. A Deep Incremental Slow Feature Analysis (D-IncSFA) network is proposed, which learns progressively abstract and global high-level features from raw data. The D-IncSFA network has the functionalities of both feature extractor and anomaly detector so anomaly detection can be completed in one step. Their approach can detect global anomaly such as crowd panic and local anomaly and is intended to be universal in order to work in different scenarios, with little human intervention and low memory and computational requirements. The paper “Multiple Deep Features Learning for Object Retrieval in Surveillance Videos” by Haiyun Guo et al. aims to address the challenge of efficiently indexing and retrieving objects of interest from large-scale surveillance videos. A multiple deep features learning approach to object retrieval in surveillance videos is proposed, which is based on the discriminative convolutional neural network (CNN). The CNN model is pre-trained on ImageNet ILSVRC12 and then fine-tuned on their dataset. To improve the retrieval performance, the deep features are encoded into short binary codes by Locality-Sensitive Hash and fused to retrieve the object of interest. Experiments on a dataset of 100k objects extracted from multi-camera surveillance videos have demonstrated good performance of the proposed approach, compared with other common visual features. The paper “A Two-layer Discriminative Model for Human Activity Recognition” by Mouna Selmi et al. studies human activity recognition with a focus on the role of local interest point features like spatio-temporal interest points. This paper presents a new approach that explicitly models the sequential aspect of activities. A support vector machine provides a vector of conditional class probabilities for each window that summarises all discriminant information that is relevant for sequence recognition. The sequence of these stochastic vectors is then fed to a hidden conditional random field for inference at the sequence level. Experiments on various human activity datasets have demonstrated that the proposed approach compares favourably with current state-of-the-art. The paper “A New Fusional Framework Combining Sparse Selection and Clustering for Key Frame Extraction” by Mengjuan Fei et al. studies key frame extraction, a type of video summarisation, which facilitates rapid browsing and efficient video indexing. This paper proposes a syncretic key frame extraction framework (SS-MIAHC) that combines sparse selection and mutual information-based agglomerative hierarchical clustering (MIAHC) to generate effective video summaries. The proposed framework overcomes issues such as information redundancy and computational complexity. The experiments conducted on two benchmark datasets demonstrate that the proposed SS-MIAHC framework is superior to conventional methods. The paper “Multi-Object Tracking using Dominant Sets” by Yonatan T. Tesfaye et al. studies multi-object tracking and addresses the challenges of identity switches and difficulties in handling long-term occlusions by formulating the tracking task as a problem of finding dominant sets in an auxiliary edge weighted graph. This is a novel approach to multi-object tracking, which has been demonstrated to have superior performance compared with several state of the art methods in experiments on three different challenging datasets. The paper “Contextualized Learning-free 3D Body Pose Estimation from 2D Body Features in Monocular Images” by Luis Unzueta et al. presents a method for 3D human body pose estimation from a monocular camera based on a learning-free hierarchical optimisation procedure and contextual information. This approach explicitly considers and preserves the relations between the 3D subject's overall scale; its depth with respect to the camera; and its configuration related to the reference floor. Thus, it can obtain more coherent reconstructions with respect to the shared 3D world, compared to other state-of-the-art approaches, efficiently and without the need for learning 2D/3D mapping models from training data. Therefore, it is not affected by data characteristic differences between training and deployment stages. The paper “‘Owl’ and ‘Lizard’: Patterns of Head Pose and Eye Pose in Driver Gaze Classification” by Lex Fridman et al. studies gaze tracking in the car through estimating head pose and eye pose from monocular video. New research questions are asked, which are answered by evaluating data drawn from an on-road study of 40 drivers. The main insight of the paper is conveyed through the analogy of an “owl” and “lizard” which describes the degree to which the eyes and the head move when shifting gaze. When the head moves a lot (“owl”), not much classification improvement is attained by estimating eye pose on top of head pose. On the other hand, when the head stays still and only the eyes move (“lizard”), classification accuracy increases significantly from adding in eye pose. The paper “Forensic Video Solution Using Facial Feature Based Synoptic Video Footage Record” by B.Yogameena et al. proposes a solution to identify a specific person quickly which is valuable in analysing incidents/crimes. The main idea of this paper is to reduce the enormous volume of video data by using an object based video synopsis. SVM is used to classify the weak and strong features. These strong features are used to recognise the person. The algorithm works well even in complicated situations such as expression changes, pose, illumination variations and even if the face is partially or fully occluded in few frames. The advantage of synoptic video helps to recognize the person who is not occluded in some other frames. Experimental results on benchmark and real time datasets demonstrate the effectiveness of the proposed algorithm. The paper “Facial Video based Detection of Physical Fatigue for Maximal Muscle Activity” by Mohammad A. Haque et al. studies video based detection of physical fatigue. This paper presents an efficient noncontact system for detecting non-localised physical fatigue from maximal muscle activity using facial videos acquired in a realistic environment with natural lighting where subjects were allowed to voluntarily move their head, change their facial expression and vary their pose. Experimental results show that the proposed system outperforms video based existing system for physical fatigue detection.

Most multi-object tracking algorithms are developed within the tracking-by-detection framework that consider the pairwise appearance similarities between detection responses or tracklets within a limited temporal window, and thus less effective in handling long-term occlusions or distinguishing spatially close targets with similar appearance in crowded scenes. In this work, we propose an algorithm that formulates the multi-object tracking task as one to exploit hierarchical dense structures on an undirected hypergraph constructed based on tracklet affinity. The dense structures indicate a group of vertices that are inter-connected with a set of hyperedges with high affinity values. The appearance and motion similarities among multiple tracklets across the spatio-temporal domain are considered globally by exploiting high-order similarities rather than pairwise ones, thereby facilitating distinguish spatially close targets with similar appearance. In addition, the hierarchical design of the optimization process helps the proposed tracking algorithm handle long-term occlusions robustly. Extensive experiments on various challenging datasets of both multi-pedestrian and multi-face tracking tasks, demonstrate that the proposed algorithm performs favorably against the state-of-the-art methods.

Learning-based face descriptors have constantly improved the face recognition performance. Compared with the hand-crafted features, learning-based features are considered to be able to exploit information with better discriminative ability for specific tasks. Motivated by the recent success of deep learning, in this paper, we extend the original shallow face descriptors to deep discriminant face features by introducing a stacked image descriptor (SID). With deep structure, more complex facial information can be extracted and the discriminant and compactness of feature representation can be improved. The SID is learned in a forward optimization way, which is computational efficient compared with deep learning. Extensive experiments on various face databases are conducted to show that SID is able to achieve high face recognition performance with compact face representation, compared with other state-of-the-art descriptors.

Robust descriptor-based subspace learning with complex data is an active topic in pattern analysis and machine intelligence. A few researches concentrate the optimal design on feature representation and metric learning. However, traditionally used features of single-type, e.g., image gradient orientations (IGOs), are deficient to characterize the complete variations in robust and discriminant subspace learning. Meanwhile, discontinuity in edge alignment and feature match are not been carefully treated in the literature. In this paper, local order constrained IGOs are exploited to generate robust features. As the difference-based filters explicitly consider the local contrasts within neighboring pixel points, the proposed features enhance the local textures and the order-based coding ability, thus discover intrinsic structure of facial images further. The multimodal features are automatically fused in the most discriminant subspace. The utilization of adaptive interaction function suppresses outliers in each dimension for robust similarity measurement and discriminant analysis. The sparsity-driven regression model is modified to adapt the classification issue of the compact feature representation. Extensive experiments are conducted by using some benchmark face data sets, e.g., of controlled and uncontrolled environments, to evaluate our new algorithm.

We present a novel Joint Online Tracking and Segmentation (JOTS) algorithm which integrates the multi-part tracking and segmentation into a unified energy optimization framework to handle the video segmentation task. The multi-part segmentation is posed as a pixel-level label assignment task with regularization according to the estimated part models, and tracking is formulated as estimating the part models based on the pixel labels, which in turn is used to refine the model. The multi-part tracking and segmentation are carried out iteratively to minimize the proposed objective function by a RANSAC-style approach. Extensive experiments on the SegTrack and SegTrack v2 databases demonstrate that the proposed algorithm performs favorably against the state-of-the-art methods.

Object detection is often conducted by object proposal generation and classification sequentially. This paper handles object detection in a superpixel oriented manner instead of the proposal oriented. Specially, this paper takes object detection as a multi-label superpixel labeling problem by minimizing an energy function. It uses the data cost term to capture the appearance, smooth cost term to encode the spatial context and label cost term to favor compact detection. The data cost is learned through a convolutional neural network and the parameters in the labeling model are learned through a structural SVM. Compared with proposal generation and classification based methods, the proposed superpixel labeling method can naturally detect objects missed by proposal generation step and capture the global image context to infer the overlapping objects. The proposed method shows its advantage in Pascal VOC and ImageNet. Notably, it performs better than the ImageNet ILSVRC2014 winner GoogLeNet (45.0% V.S. 43.9% in mAP) with much shallower and fewer CNNs.

Pose and expression normalization is a crucial step to recover the canonical view of faces under arbitrary conditions, so as to improve the face recognition performance. An ideal normalization method is desired to be automatic, database independent and high-fidelity, where the face appearance should be preserved with little artifact and information loss. However, most normalization methods fail to satisfy one or more of the goals. In this paper, we propose a High-fidelity Pose and Expression Normalization (HPEN) method with 3D Morphable Model (3DMM) which can automatically generate a natural face image in frontal pose and neutral expression. Specifically, we firstly make a landmark marching assumption to describe the non-correspondence between 2D and 3D landmarks caused by pose variations and propose a pose adaptive 3DMM fitting algorithm. Secondly, we mesh the whole image into a 3D object and eliminate the pose and expression variations using an identity preserving 3D transformation. Finally, we propose an inpainting method based on Possion Editing to fill the invisible region caused by self occlusion. Extensive experiments on Multi-PIE and LFW demonstrate that the proposed method significantly improves face recognition performance and outperforms state-of-the-art methods in both constrained and unconstrained environments.

Deep learning methods are powerful tools but often suffer from expensive computation and limited flexibility. An alternative is to combine light-weight models with deep representations. As successful cases exist in several visual problems, a unified framework is absent. In this paper, we revisit two widely used approaches in computer vision, namely filtered channel features and Convolutional Neural Networks (CNN), and absorb merits from both by proposing an integrated method called Convolutional Channel Features (CCF). CCF transfers low-level features from pre-trained CNN models to feed the boosting forest model. With the combination of CNN features and boosting forest, CCF benefits from the richer capacity in feature representation compared with channel features, as well as lower cost in computation and storage compared with end-to-end CNN methods. We show that CCF serves as a good way of tailoring pre-trained CNN models to diverse tasks without fine-tuning the whole network to each task by achieving state-of-the-art performances in pedestrian detection, face detection, edge detection and object proposal generation.

Recent advances in visual tracking have focused on handling deformations and occlusions using the part-based appearance model. However, it remains a challenge to come up with a reliable target representation using local parts, and hence existing trackers continue to face drifting problems. To deal with this challenge, we propose a robust online model, formulating the tracking task as a problem of identifying Temporally Coherent Part (TCP) clusters. Specifically, we pose the TCP clusters identification task as a dense neighborhoods searching problem using a relational hyper graph in which the relationship among multiple temporal local parts is encoded as the affinity value of a hyper edge connecting them. Such high-order relations ships among multiple local parts across the temporal domain make our tracker more robust towards deformations and occlusions. Extensive experiments on various challenging video sequences demonstrate that our TCP-based method performs better than the state-of-the-art methods.

The detection of moving objects in videos is very important in many video processing applications, and background modeling is often an indispensable process to achieve this goal. Most of the traditional background modeling methods utilize color or texture information. However, color information is sensitive to illumination variations and texture information cannot be utilized to separate smooth foreground from smooth background in most cases. Achieving good performance in terms of high foreground detection accuracy and low computational cost is also challenging. In this paper, we propose a new integration framework of texture and color information for background modeling, in which the foreground decision equation includes three parts (one part for color information, one part for texture information, and the left part for the integration of color and texture information). This framework is able to combine the advantages of texture and color features while inhibiting their disadvantages as well. Moreover, we propose a block-based method to accelerate the background modeling. In particular, in the texture information modeling process, a single histogram model is established for each block whose bins indicate the occurrence probabilities of different patterns, which is different from the traditional multihistogram model for block-based background modeling, and then dominant background patterns are selected to calculate the background likelihood of new coming blocks. Dynamic background and multimodal problems can be handled through this technique. To evaluate the foreground detection performance reasonably, a new quality measure is proposed. Extensive experiments on various challenging videos validate the effectiveness of the proposed method over state-of-the-art methods.

Video synopsis or condensation is a smart solution for fast video browsing and storage. However, most of the existing methods work offline, where two main phases are required. The first phase is to prepare tubes and background images. The second phase is to rearrange tubes and stitch them into backgrounds. However, with a long video sequence, the first phase is memory consuming for data storage, and the second phase is computationally expensive to rearrange all tubes simultaneously. To overcome these problems, we propose a high-performance video condensation system based on an online content-aware framework. The online framework transforms the optimization problem of tube rearrangement into a stepwise optimization problem. Therefore, it can condense video with much less memory and higher speed than the offline framework. With the aid of this transformation, the proposed system can process input videos and produce condensed videos simultaneously. Thus it is suitable for real-time endless surveillance videos. Meanwhile, the online mechanism allows users to directly visit the condensation video that has been generated. Moreover, the content-aware mechanism makes the proposed system able to automatically determine the duration of a condensed video. Finally, the proposed system uses Graphic Processing Unit (GPU) and multicore techniques to improve the speed. Extensive experiments that validate the high efficiency of the system are presented.

Face antispoofing is important to practical face recognition systems. In previous works, a generic antispoofing classifier is trained to detect spoofing attacks on all subjects. However, due to the individual differences among subjects, the generic classifier cannot generalize well to all subjects. In this paper, we propose a person-specific face antispoofing approach. It recognizes spoofing attacks using a classifier specifically trained for each subject, which dismisses the interferences among subjects. Moreover, considering the scarce or void fake samples for training, we propose a subject domain adaptation method to synthesize virtual features, which makes it tractable to train well-performed individual face antispoofing classifiers. The extensive experiments on two challenging data sets: 1) CASIA and 2) REPLAY-ATTACK demonstrate the prospect of the proposed approach.

Multi-target tracking is an interesting but challenging task in computer vision field. Most previous data association based methods merely consider the relationships (e.g. appearance and motion pattern similarities) between detections in local limited temporal domain, leading to their difficulties in handling long-term occlusion and distinguishing the spatially close targets with similar appearance in crowded scenes. In this paper, a novel data association approach based on undirected hierarchical relation hypergraph is proposed, which formulates the tracking task as a hierarchical dense neighborhoods searching problem on the dynamically constructed undirected affinity graph. The relationships between different detections across the spatiotemporal domain are considered in a high-order way, which makes the tracker robust to the spatially close targets with similar appearance. Meanwhile, the hierarchical design of the optimization process fuels our tracker to long-term occlusion with more robustness. Extensive experiments on various challenging datasets (i.e. PETS2009 dataset, ParkingLot), including both low and high density sequences, demonstrate that the proposed method performs favorably against the state-of-the-art methods.

This paper solves the speed bottleneck of deformable part model (DPM), while maintaining the accuracy in detection on challenging datasets. Three prohibitive steps in cascade version of DPM are accelerated, including 2D correlation between root filter and feature map, cascade part pruning and HOG feature extraction. For 2D correlation, the root filter is constrained to be low rank, so that 2D correlation can be calculated by more efficient linear combination of 1D correlations. A proximal gradient algorithm is adopted to progressively learn the low rank filter in a discriminative manner. For cascade part pruning, neighborhood aware cascade is proposed to capture the dependence in neighborhood regions for aggressive pruning. Instead of explicit computation of part scores, hypotheses can be pruned by scores of neighborhoods under the first order approximation. For HOG feature extraction, look-up tables are constructed to replace expensive calculations of orientation partition and magnitude with simpler matrix index operations. Extensive experiments show that (a) the proposed method is 4 times faster than the current fastest DPM method with similar accuracy on Pascal VOC, (b) the proposed method achieves state-of-the-art accuracy on pedestrian and face detection task with frame-rate speed.

Many efforts have been made in recent years to tackle the unconstrained face recognition challenge. For the benchmark of this challenge, the Labeled Faces in theWild (LFW) database has been widely used. However, the standard LFW protocol is very limited, with only 3,000 genuine and 3,000 impostor matches for classification. Today a 97% accuracy can be achieved with this benchmark, remaining a very limited room for algorithm development. However, we argue that this accuracy may be too optimistic because the underlying false accept rate may still be high (e.g. 3%). Furthermore, performance evaluation at low FARs is not statistically sound by the standard protocol due to the limited number of impostor matches. Thereby we develop a new benchmark protocol to fully exploit all the 13,233 LFW face images for large-scale unconstrained face recognition evaluation under both verification and open-set identification scenarios, with a focus at low FARs. Based on the new benchmark, we evaluate 21 face recognition approaches by combining 3 kinds of features and 7 learning algorithms. The benchmark results show that the best algorithm achieves 41.66% verification rates at FAR=0.1%, and 18.07% open-set identification rates at rank 1 and FAR=1%. Accordingly we conclude that the large-scale unconstrained face recognition problem is still largely unresolved, thus further attention and effort is needed in developing effective feature representations and learning algorithms. We thereby release a benchmark tool to advance research in this field.

Face detection has drawn much attention in recent decades since the seminal work by Viola and Jones. While many subsequences have improved the work with more powerful learning algorithms, the feature representation used for face detection still can't meet the demand for effectively and efficiently handling faces with large appearance variance in the wild. To solve this bottleneck, we borrow the concept of channel features to the face detection domain, which extends the image channel to diverse types like gradient magnitude and oriented gradient histograms and therefore encodes rich information in a simple form. We adopt a novel variant called aggregate channel features, make a full exploration of feature design, and discover a multi-scale version of features with better performance. To deal with poses of faces in the wild, we propose a multi-view detection approach featuring score re-ranking and detection adjustment. Following the learning pipelines in Viola-Jones framework, the multi-view face detector using aggregate channel features shows competitive performance against state-of-the-art algorithms on AFW and FDDB test-sets, while runs at 42 FPS on VGA images.

In recent years, large-scale video search and mining has been an active research area. Exploring the trajectory of pedestrian of interest in non-overlapping multi-camera network, namely the trajectory mining, is very useful for visual surveillance and criminal investigation. The trajectory mentioned in our work describes the transition of pedestrian among cameras from a macroscopic perspective which is different from the concept in conventional tracking field. In this paper, we collect a database called TMin to promote research and development of trajectory mining. This release of Version 1 contains 1680 images from 30 subjects, all the images are extracted from 6 surveillance videos over two hours, and each subject appears in at least two different cameras. We describe the apparatuses, environments and procedure of the data collection and present baseline performance on the TMin database.

LBP is an effective descriptor for face recognition. LBP encodes the ordinal relationship between the neighborhood samplings and the central one to obtain robust face representation. However, additional information like the difference among neighboring pixels, which may be helpful for face recognition, is ignored. On the other hand, gradient information which enhances the edge response and suppresses the external noise like illumination variation, is usually useful for face recognition. In this paper, we propose a novel face descriptor, namely local gradient order pattern (LGOP), taking into account the ordinal relationship of gradient responses in local region to obtain robust face representation. After pattern encoding, a 2-D histogram is consequently adopted to calculate the occurrence frequency of different patterns and multi-scale histogram features are extracted to represent the face image. We further adopt whitened principal component analysis (WPCA) to reduce the feature dimensionality and improve the computational efficiency. Extensive experiments on FERET, CAS-PEAL and LFW validates the effectiveness of LGOP for both constrained and unconstrained face recognition problems.

Due to illumination changes, partial occlusions, and object scale differences, person re-identification over disjoint camera views becomes a challenging problem. To address this problem, a variety of image representations have been put forward. In this paper, the illumination invariance and distinctiveness of different color models including the proposed color model are firstly evaluated. Since color distribution is robust to image scales and partial occlusions, color distributions based on different color models are then calculated and fused in the stage of feature extraction. Different color models obtain robustness to different types of illumination and thus fusing them can compensate each other and contribute to better performance. In the stage of feature matching, a weighted KISSME is presented to learn a better distance metric than the original KISSME. Experimental results demonstrate its feasibility and effectiveness. Finally, image pairs are matched based on the learned distance metric. Experiments conducted on two public benchmark datasets (VIPeR and PRID 450S) show that the proposed algorithm outperforms the state-of-the-art methods.

Various hand-crafted features and metric learning methods prevail in the field of person re-identification. Compared to these methods, this paper proposes a more general way that can learn a similarity metric from image pixels directly. By using a "siamese" deep neural network, the proposed method can jointly learn the color feature, texture feature and metric in a unified framework. The network has a symmetry structure with two sub-networks which are connected by a cosine layer. Each sub network includes two convolutional layers and a full connected layer. To deal with the big variations of person images, binomial deviance is used to evaluate the cost between similarities and labels, which is proved to be robust to outliers. Experiments on VIPeR illustrate the superior performance of our method and a cross database experiment also shows its good generalization.

3D Morph able Model (3DMM) has been widely used in face analysis for many years. The most challenging part of 3DMM is to find the correspondences between 3D points and 2D pixels. Existing methods only use key points, edges, specular highlights and image pixels to complete the task, which are not accurate or robust. This paper proposes a new algorithm called Sparse SIFT Flow (SSF) to improve the reconstruction accuracy. We mark a set of salient points to control the shape of facial components and use SSF to find their corresponding pixels on the input image. We also incorporate SSF into Multi-Features Framework to construct a robust 3DMM fitting algorithm. Compared with the state-of-the art, our approach significantly improves the fitting results in facial component area.

Local feature descriptor is an important module for face recognition and those like Gabor and local binary patterns (LBP) have proven effective face descriptors. Traditionally, the form of such local descriptors is predefined in a handcrafted way. In this paper, we propose a method to learn a discriminant face descriptor (DFD) in a data-driven way. The idea is to learn the most discriminant local features that minimize the difference of the features between images of the same person and maximize that between images from different people. In particular, we propose to enhance the discriminative ability of face representation in three aspects. First, the discriminant image filters are learned. Second, the optimal neighborhood sampling strategy is soft determined. Third, the dominant patterns are statistically constructed. Discriminative learning is incorporated to extract effective and robust features. We further apply the proposed method to the heterogeneous (cross-modality) face recognition problem and learn DFD in a coupled way (coupled DFD or C-DFD) to reduce the gap between features of heterogeneous face images to improve the performance of this challenging problem. Extensive experiments on FERET, CAS-PEAL-R1, LFW, and HFB face databases validate the effectiveness of the proposed DFD learning on both homogeneous and heterogeneous face recognition problems. The DFD improves POEM and LQP by about 4.5 percent on LFW database and the C-DFD enhances the heterogeneous face recognition performance of LBP by over 25 percent.

Face recognition (FR) systems in real-world applications need to deal with a wide range of interferences, such as occlusions and disguises in face images. Compared with other forms of interferences such as nonuniform illumination and pose changes, face with occlusions has not attracted enough attention yet. A novel approach, coined dynamic image-to-class warping (DICW), is proposed in this work to deal with this challenge in FR. The face consists of the forehead, eyes, nose, mouth, and chin in a natural order and this order does not change despite occlusions. Thus, a face image is partitioned into patches, which are then concatenated in the raster scan order to form an ordered sequence. Considering this order information, DICW computes the image-to-class distance between a query face and those of an enrolled subject by finding the optimal alignment between the query sequence and all sequences of that subject along both the time dimension and within-class dimension. Unlike most existing methods, our method is able to deal with occlusions which exist in both gallery and probe images. Extensive experiments on public face databases with various types of occlusions have confirmed the effectiveness of the proposed method.

While some efforts have been paid to handle deformation and occlusion in visual tracking, they are still great challenges. In this paper, a dynamic graph-based tracker (DGT) is proposed to address these two challenges in a unified framework. In the dynamic target graph, nodes are the target local parts encoding appearance information, and edges are the interactions between nodes encoding inner geometric structure information. This graph representation provides much more information for tracking in the presence of deformation and occlusion. The target tracking is then formulated as tracking this dynamic undirected graph, which is also a matching problem between the target graph and the candidate graph. The local parts within the candidate graph are separated from the background with Markov random field, and spectral clustering is used to solve the graph matching. The final target state is determined through a weighted voting procedure according to the reliability of part correspondence, and refined with recourse to a foreground/background segmentation. An effective online updating mechanism is proposed to update the model, allowing DGT to robustly adapt to variations of target structure. Experimental results show improved performance over several state-of-the-art trackers, in various challenging scenarios.

Visual tracking is an important but challenging problem in the computer vision field. In the real world, the appearances of the target and its surroundings change continuously over space and time, which provides effective information to track the target robustly. However, enough attention has not been paid to the spatio-temporal appearance information in previous works. In this paper, a robust spatio-temporal context model based tracker is presented to complete the tracking task in unconstrained environments. The tracker is constructed with temporal and spatial appearance context models. The temporal appearance context model captures the historical appearance of the target to prevent the tracker from drifting to the background in a long-term tracking. The spatial appearance context model integrates contributors to build a supporting field. The contributors are the patches with the same size of the target at the key-points automatically discovered around the target. The constructed supporting field provides much more information than the appearance of the target itself, and thus, ensures the robustness of the tracker in complex environments. Extensive experiments on various challenging databases validate the superiority of our tracker over other state-of-the-art trackers.

The serious performance decline with decreasing resolution is the major bottleneck for current pedestrian detection techniques. In this paper, we take pedestrian detection in different resolutions as different but related problems, and propose a Multi-Task model to jointly consider their commonness and differences. The model contains resolution aware transformations to map pedestrians in different resolutions to a common space, where a shared detector is constructed to distinguish pedestrians from background. For model learning, we present a coordinate descent procedure to learn the resolution aware transformations and deformable part model (DPM) based detector iteratively. In traffic scenes, there are many false positives located around vehicles, therefore, we further build a context model to suppress them according to the pedestrian-vehicle relationship. The context model can be learned automatically even when the vehicle annotations are not available. Our method reduces the mean miss rate to 60% for pedestrians taller than 30 pixels on the Caltech Pedestrian Benchmark, which noticeably outperforms previous state-of-the-art (71%).

Most existing pose robust methods are too computational complex to meet practical applications and their performance under unconstrained environments are rarely evaluated. In this paper, we propose a novel method for pose robust face recognition towards practical applications, which is fast, pose robust and can work well under unconstrained environments. Firstly, a 3D deformable model is built and a fast 3D model fitting algorithm is proposed to estimate the pose of face image. Secondly, a group of Gabor filters are transformed according to the pose and shape of face image for feature extraction. Finally, PCA is applied on the pose adaptive Gabor features to remove the redundances and Cosine metric is used to evaluate the similarity. The proposed method has three advantages: (1) The pose correction is applied in the filter space rather than image space, which makes our method less affected by the precision of the 3D model, (2) By combining the holistic pose transformation and local Gabor filtering, the final feature is robust to pose and other negative factors in face recognition, (3) The 3D structure and facial symmetry are successfully used to deal with self-occlusion. Extensive experiments on FERET and PIE show the proposed method outperforms state-of-the-art methods significantly, meanwhile, the method works well on LFW.

Face recognition is confronted with situations in which face images are captured in various modalities, such as the visual modality, the near infrared modality, and the sketch modality. This is known as heterogeneous face recognition. To solve this problem, we propose a new method called discriminative spectral regression (DSR). The DSR maps heterogeneous face images into a common discriminative subspace in which robust classification can be achieved. In the proposed method, the subspace learning problem is transformed into a least squares problem. Different mappings should map heterogeneous images from the same class close to each other, while images from different classes should be separated as far as possible. To realize this, we introduce two novel regularization terms, which reflect the category relationships among data, into the least squares approach. Experiments conducted on two heterogeneous face databases validate the superiority of the proposed method over the previous methods.

Although numerous online learning strategies have been proposed to handle the appearance variation in visual tracking, the existing methods just perform well in certain cases since they lack effective appearance learning mechanism. In this paper, a joint model tracker (JMT) is presented, which consists of a generative model based on Multiple Subspaces and a discriminative model based on improved Multiple Instance Boosting (MIBoosting). The generative model utilizes a series of local constructed subspaces to update the Multiple Subspaces model and considers the energy dissipation of dimension reduction in updating step. The discriminative model adopts the Gaussian Mixture Model (GMM) to estimate the posterior probability of the likelihood function. These two parts supervise each other to update in multiple instance way which helps our tracker recover from drift. Extensive experiments on various databases validate the effectiveness of our proposed method over other state-of-the-art trackers.

With the increasing amount of surveillance data, moving object segmentation in the compressed domain has drawn broad attention from both academy and industry. In this paper, we propose a novel moving object segmentation method towards H.264 compressed surveillance videos. First, the motion vectors (MV) are accumulated and filtered to achieve reliable motion information. Second, considering the spatial and temporal correlations among adjacent blocks, spatio-temporal Local Binary Pattern (LBP) features of MVs are extracted to obtain coarse and initial object regions. Finally, a coarse-to-fine segmentation algorithm of boundary modification is conducted based on the DCT coefficients. The experimental results validate that the proposed method not only can extract fairly accurate objects in compressed video, but also has a relatively low computational complexity.

People counting is one of the key components in video surveillance applications, however, due to occlusion, illumination, color and texture variation, the problem is far from being solved. Different from traditional visible camera based systems, we construct a novel system that uses vertical Kinect sensor for people counting, where the depth information is used to remove the affect of the appearance variation. Since the head is always closer to the Kinect sensor than other parts of the body, people counting task equals to find the suitable local minimum regions. According to the particularity of the depth map, we propose a novel unsupervised water filling method that can find these regions with the property of robustness, locality and scale-invariance. Experimental comparisons with mean shift and random forest on two databases validate the superiority of our water filling algorithm in people counting.

Explosive growth of surveillance video data presents formidable challenges to its browsing, retrieval and storage. Video synopsis, an innovation proposed by Peleg and his colleagues, is aimed for fast browsing by shortening the video into a synopsis while keeping activities in video captured by a camera. However, the current techniques are offline methods requiring that all the video data be ready for the processing, and are expensive in time and space. In this paper, we propose an online and efficient solution, and its supporting algorithms to overcome the problems. The method adopts an online content-aware approach in a step-wise manner, hence applicable to endless video, with less computational cost. Moreover, we propose a novel tracking method, called sticky tracking, to achieve high-quality visualization. The system can achieve a faster-than-real-time speed with a multi-core CPU implementation. The advantages are demonstrated by extensive experiments with a wide variety of videos. The proposed solution and algorithms could be integrated with surveillance cameras, and impact the way that surveillance videos are recorded.

Local binary pattern (LBP) and its variants are effective descriptors for face recognition. The traditional LBP like features are extracted based on the original pixel or patch values of images. In this paper, we propose to learn the discriminative image filter to improve the discriminant power of the LBP like feature. The basic idea is after the image filtering with the learned filter, the difference of pixel difference vectors (PDVs) between the images from the same person is consistent and the difference between the images from different persons is enlarged. In this way, the LBP like features extracted from the filtered images are considered to be more discriminant than those extracted from the original images. Moreover, a coupled discriminant image filters learning method is proposed to deal with the heterogenous face images matching problem by reducing the feature gap between the heterogeneous images. Experiments on FERET, FRGC and a VIS-NIR heterogeneous face databases validate the effectiveness of our proposed image filter learning method combined with LBP like features.

Recent state-of-the-art algorithms have achieved good performance on normal pedestrian detection tasks. However, pedestrian detection in crowded scenes is still challenging due to the significant appearance variation caused by heavy occlusions and complex spatial interactions. In this paper we propose a unified probabilistic framework to globally describe multiple pedestrians in crowded scenes in terms of appearance and spatial interaction. We utilize a mixture model, where every pedestrian is assumed in a special subclass and described by the sub-model. Scores of pedestrian parts are used to represent appearance and quadratic kernel is used to represent relative spatial interaction. For efficient inference, multi-pedestrian detection is modeled as a MAP problem and we utilize greedy algorithm to get an approximation. For discriminative parameter learning, we formulate it as a learning to rank problem, and propose Latent Rank SVM for learning from weakly labeled data. Experiments on various databases validate the effectiveness of the proposed approach.

Liveness detection is an indispensable guarantee for reliable face recognition, which has recently received enormous attention. In this paper we propose three scenic clues, which are non-rigid motion, face-background consistency and imaging banding effect, to conduct accurate and efficient face liveness detection. Non-rigid motion clue indicates the facial motions that a genuine face can exhibit such as blinking, and a low rank matrix decomposition based image alignment approach is designed to extract this non-rigid motion. Face-background consistency clue believes that the motion of face and background has high consistency for fake facial photos while low consistency for genuine faces, and this consistency can serve as an efficient liveness clue which is explored by GMM based motion detection method. Image banding effect reflects the imaging quality defects introduced in the fake face reproduction, which can be detected by wavelet decomposition. By fusing these three clues, we thoroughly explore sufficient clues for liveness detection. The proposed face liveness detection method achieves 100% accuracy on Idiap print-attack database and the best performance on self-collected face anti-spoofing database.

This letter addresses the problem of face detection in multispectral illuminations. Face detection in visible images has been well addressed based on the large scale training samples. For the recently emerging multispectral face biometrics, however, the face data is scarce and expensive to collect, and it is usually short of face samples to train an accurate face detector. In this letter, we propose to tackle the issue of multispectral face detection by combining existing large scale visible face images and a few multispectral face images. We cast the problem of face detection across spectrum into the transfer learning framework and try to learn the robust multispectral face detector by exploring relevant knowledge from visible data domain. Specifically, a novel Regularized Transfer Boosting algorithm named R-TrBoost is proposed, with features of weighted loss objective and manifold regularization. Experiments are performed with face images of two spectrums, 850 nm and 365 nm, and the results show significant improvement on multispectral face detection using the proposed algorithm.

Coupled space learning is an effective framework for heterogeneous face recognition. In this paper, we propose a novel coupled discriminant analysis method to improve the heterogeneous face recognition performance. There are two main advantages of the proposed method. First, all samples from different modalities are used to represent the coupled projections, so that sufficient discriminative information could be extracted. Second, the locality information in kernel space is incorporated into the coupled discriminant analysis as a constraint to improve the generalization ability. In particular, two implementations of locality constraint in kernel space (LCKS)-based coupled discriminant analysis methods, namely LCKS-coupled discriminant analysis (LCKS-CDA) and LCKS-coupled spectral regression (LCKS-CSR), are presented. Extensive experiments on three cases of heterogeneous face matching (high versus low image resolution, digital photo versus video image, and visible light versus near infrared) validate the efficacy of the proposed method.

Face recognition from low-resolution images is a common yet challenging case in real applications. Since the high-frequency information is lost in low-resolution images, it is necessary to explore robust information in the low frequency domain. In this paper, we propose an effective local frequency descriptor (LFD) for low resolution face recognition, by building upon the ideas behind local phase quantization (LPQ) and exploring both blur-invariant magnitude and phase information in the low frequency domain. The proposed descriptor is more descriptive than LPQ with more comprehensive information. In addition, a statistical uniform pattern definition method is introduced to improve the efficiency of the proposed descriptor. Experimental results on FERET and a real video database show that LFD is effective and robust for low-resolution face recognition.

Existing face liveness detection algorithms adopt behavioural challenge-response methods that require user cooperation. To be verified live, users are expected to obey some user unfriendly requirement. In this paper, we present a multispectral face liveness detection method, which is user cooperation free. Moreover, the system is adaptive to various user-system distances. Using the Lambertian model, we analyze multispectral properties of human skin versus non-skin, and the discriminative wavelengths are then chosen. Reflectance data of genuine and fake faces at multi-distances are selected to form a training set. An SVM classifier is trained to learn the multispectral distribution for a final Genuine-or-Fake classification. Compared with previous works, the proposed method has the following advantages: (a) The requirement on the users' cooperation is no longer needed, making the liveness detection user friendly and fast. (b) The system can work without restricted distance requirement from the target being analyzed. Experiments are conducted on genuine versus planar face data, and genuine versus mask face data. Furthermore a comparison with the visible challenge-response liveness detection method is also given. The experimental results clearly demonstrate the superiority of our method over previous systems.

Spoofing identities using photographs is one of the most common techniques to attack 2-D face recognition systems. There seems to exist no comparative studies of different techniques using the same protocols and data. The motivation behind this competition is to compare the performance of different state-of-the-art algorithms on the same database using a unique evaluation method. Six different teams from universities around the world have participated in the contest. Use of one or multiple techniques from motion, texture analysis and liveness detection appears to be the common trend in this competition. Most of the algorithms are able to clearly separate spoof attempts from real accesses. The results suggest the investigation of more complex attacks.

Linear discriminant analysis with nearest neighborhood classifier (LDA + NN) has been commonly used in face recognition, but it often confronts with two problems in real applications: (1) it cannot incrementally deal with the information of training instances; (2) it cannot achieve fast search against large scale gallery set. In this paper, we use incremental LDA (ILDA) and hashing based search method to deal with these two problems. Firstly two incremental LDA algorithms are proposed under spectral regression framework, namely exact incremental spectral regression discriminant analysis (EI-SRDA) and approximate incremental spectral regression discriminant analysis (AI-SRDA). Secondly we propose a similarity hashing algorithm of sub-linear complexity to achieve quick recognition from large gallery set. Experiments on FRGC and self-collected 100,000 faces database show the effective of our methods.

Eye localization is an important part in face recognition system, because its precision closely affects the performance of face recognition. Although various methods have already achieved high precision on the face images with high quality, their precision will drop on low quality images. In this paper, we propose a robust eye localization method for low quality face images to improve the eye detection rate and localization precision. First, we propose a probabilistic cascade (P-Cascade) framework, in which we reformulate the traditional cascade classifier in a probabilistic way. The P-Cascade can give chance to each image patch contributing to the final result, regardless the patch is accepted or rejected by the cascade. Second, we propose two extensions to further improve the robustness and precision in the P-Cascade framework. There are: (1) extending feature set, and (2) stacking two classifiers in multiple scales. Extensive experiments on JAFFE, BioID, LFW and a self-collected video surveillance database show that our method is comparable to state-of-the-art methods on high quality images and can work well on low quality images. This work supplies a solid base for face recognition applications under unconstrained or surveillance environments.

Low resolution (LR) is an important issue when handling real world face recognition problems. The performance of traditional recognition algorithms will drop drastically due to the loss of facial texture information in original high resolution (HR) images. To address this problem, in this paper we propose an effective approach named Simultaneous Discriminant Analysis (SDA). SDA learns two mappings from LR and HR images respectively to a common subspace where discrimination property is maximized. In SDA, (1) the data gap between LR and HR is reduced by mapping into a common space; and (2) the mapping is designed for preserving most discriminative information. After that, the conventional classification method is applied in the common space for final decision. Extensive experiments are conducted on both FERET and Multi-PIE, and the results clearly show the superiority of the proposed SDA over state-of-the-art methods.

Information jointly contained in image space, scale and orientation domains can provide rich important clues not seen in either individual of these domains. The position, spatial frequency and orientation selectivity properties are believed to have an important role in visual perception. This paper proposes a novel face representation and recognition approach by exploring information jointly in image space, scale and orientation domains. Specifically, the face image is first decomposed into different scale and orientation responses by convolving multiscale and multiorientation Gabor filters. Second, local binary pattern analysis is used to describe the neighboring relationship not only in image space, but also in different scale and orientation responses. This way, information from different domains is explored to give a good face representation for recognition. Discriminant classification is then performed based upon weighted histogram intersection or conditional mutual information with linear discriminant analysis techniques. Extensive experimental results on FERET, AR, and FRGC ver 2.0 databases show the significant advantages of the proposed method over the existing ones.

Moving cast shadow removal is an important yet difficult problem in video analysis and applications. This paper presents a novel algorithm for detection of moving cast shadows, that based on a local texture descriptor called Scale Invariant Local Ternary Pattern (SILTP). An assumption is made that the texture properties of cast shadows bears similar patterns to those of the background beneath them. The likelihood of cast shadows is derived using information in both color and texture. An online learning scheme is employed to update the shadow model adaptively. Finally, the posterior probability of cast shadow region is formulated by further incorporating prior contextual constrains using a Markov Random Field (MRF) model. The optimal solution is found using graph cuts. Experimental results tested on various scenes demonstrate the robustness of the algorithm.

Face recognition algorithms need to deal with variable lighting conditions. Near infrared (NIR) image based face recognition technology has been proposed to effectively overcome this difficulty. However, it requires that the enrolled face images be captured using NIR images whereas many applications require visual (VIS) images for enrollment templates. To take advantage of NIR face images for illumination-invariant face recognition and allow the use of VIS face images for enrollment, we encounter a new face image pattern recognition problem, that is, heterogeneous face matching between NIR versus VIS faces. In this paper, we present a subspace learning framework named Coupled Spectral Regression (CSR) to solve this challenge problem of coupling the two types of face images and matching between them. CSR first models the properties of different types of data separately and then learns two associated projections to project heterogeneous data (e.g. VIS and NIR) respectively into a discriminative common subspace in which classification is finally performed. Compared to other existing methods, CSR is computational efficient, benefiting from the efficiency of spectral regression and has better generalization performance. Experimental results on VIS-NIR face database show that the proposed CSR method significantly outperforms the existing methods.

In this paper, we propose a new approach for face shape recovery from a single image. A single near infrared (NIR) image is used as the input, and a mapping from the NIR tensor space to 3D tensor space, learned by using statistical learning, is used for the shape recovery. In the learning phase, the two tensor models are constructed for NIR and 3D images respectively, and a canonical correlation analysis (CCA) based multi-variate mapping from NIR to 3D faces is learned from a given training set of NIR-3D face pairs. In the reconstruction phase, given an NIR face image, the depth map is computed directly using the learned mapping with the help of tensor models. Experimental results are provided to evaluate the accuracy and speed of the method. The work provides a practical solution for reliable and fast shape recovery and modeling of 3D objects.

Linear Discriminant Analysis (LDA) has been a popular method for feature extracting and face recognition. As a supervised method, it requires manually labeled samples for training, while making labeled samples is a time consuming and exhausting work. A semi-supervised LDA (SDA [3]) has been proposed recently to enable training of LDA with partially labeled samples. In this paper, we first reformulate supervised LDA based on the normalized perspective of LDA. Then we show that such a reformulation is powerful for semi-supervised learning of LDA. We call this approach Normalized LDA, which uses total diversity to normalize intra-class diversity and aims to find projection directions that minimize normalized intra-class diversity. Although the Normalized LDA is identical to LDA in the supervised situation, a semi-supervised approach can be easily incorporated into its framework to make use of unlabeled samples to improve the performance in the learned subspace. Moreover, different with SDA which uses unlabeled samples to preserve neighboring relations, unlabeled samples in the Normalized LDA are used for a more accurate estimation of data space. Experiments of face recognition on the FRGC version 2 database and CMU PIE database demonstrate that the Normalized LDA outperforms SDA.

Active shape model (ASM) statistically represents a shape by a set of well-defined landmark points and models object variations using principal component analysis (PCA). However, the extracted shape contour modeled by PCA is still unsmooth when the shape has a large variation compared with the mean shape. In this paper, we propose a regularized ASM (R-ASM) model for shape alignment. During training stage, we present a regularized shape subspace on which image smoothness constraint is imposed, such that the learned components to model shape variations should not only minimize reconstruction error but also obey smoothness principle. During searching stage, a coarse-to-fine parameter adjustment strategy is performed under Bayesian inference. It makes a desired shape smoother and more robust to local noise. Lastly, an inner shape is introduced to further regularize search results. Experiments on face alignment demonstrate the efficiency and effectiveness of our proposed approach.

This paper presents a novel face representation and recognition approach. The face image is first decomposed by multi-scale and multi-orientation Gabor filters and local binary pattern (LBP) analysis is then applied on the derived Gabor magnitude responses. Different from (W.C. Zhang et al., 2005), the present method not only describes the neighboring relationship in spatial domain, but also exploit those between different scales (frequency) and orientations. Specifically, we first reformulate the Gabor magnitude responses as a 3rd-order volume and then apply LBP analysis on three orthogonal planes of the Gabor volume, named GV-LBP-TOP in short, in a hope to encode sufficient information for face representation. Further, a computationally effective version, E-GV-LBP, is proposed to depict the neighboring changes in spatial, frequency and orientation domains simultaneously. Finally, the weighted histogram intersection metric is utilized to measure the dissimilarity of faces. Experimental results on FERET and FRGC ver 2.0 databases show the significant advantages of the proposed method.

In recent years, 3D face recognition has obtained much attention. Using 2D face image as probe and 3D face data as gallery is an alternative method to deal with computation complexity, expensive equipment and fussy pretreatment in 3D face recognition systems. In this paper we propose a learning based 2D-3D face matching method using the CCA to learn the mapping between 2D face image and 3D face data. This method makes it possible to match the on-site 2D face image with enrolled 3D face data. Our 2D-3D face matching method decreased the computation complexity drastically compared to the conventional 3D-3D face matching while keeping relative high recognition rate. Furthermore, to simplify the mapping between 2D face image and 3D face data, a patch based strategy is proposed to boost the accuracy of matching. And the kernel method is also evaluated to reveal the non-linear relationship. The experiment results show that CCA based method has good performance and patch based method has significant improvement compared to the holistic method.