Text-to-image diffusion models (T2I) have demonstrated unprecedented capabilities in creating realistic and aesthetic images. On the contrary, text-to-video diffusion models (T2V) still lag far behind in frame quality and text alignment, owing to insufficient quality and quantity of training videos. In this paper, we introduce VideoElevator, a training-free and plug-and-play method, which elevates the performance of T2V using superior capabilities of T2I. Different from conventional T2V sampling (i.e., temporal and spatial modeling), VideoElevator explicitly decomposes each sampling step into temporal motion refining and spatial quality elevating. Specifically, temporal motion refining uses encapsulated T2V to enhance temporal consistency, followed by inverting to the noise distribution required by T2I. Then, spatial quality elevating harnesses inflated T2I to directly predict less noisy latent, adding more photo-realistic details. We have conducted experiments in extensive prompts under the combination of various T2V and T2I. The results show that VideoElevator not only improves the performance of T2V baselines with foundational T2I, but also facilitates stylistic video synthesis with personalized T2I. Please watch all videos in supplementary materials for better view.

MetaDesigner introduces a transformative framework for artistic typography synthesis, powered by Large Language Models (LLMs) and grounded in a user-centric design paradigm. Its foundation is a multi-agent system comprising the Pipeline, Glyph, and Texture agents, which collectively orchestrate the creation of customizable WordArt, ranging from semantic enhancements to intricate textural elements. A central feedback mechanism leverages insights from both multimodal models and user evaluations, enabling iterative refinement of design parameters. Through this iterative process, MetaDesigner dynamically adjusts hyperparameters to align with user-defined stylistic and thematic preferences, consistently delivering WordArt that excels in visual quality and contextual resonance. Empirical evaluations underscore the system's versatility and effectiveness across diverse WordArt applications, yielding outputs that are both aesthetically compelling and context-sensitive.

Accurate 3D Human Pose Estimation (HPE) in video sequences demands both precision and a robust architectural framework. Building upon the recent success of transformers in computer vision, we introduce the Refined Temporal Pyramidal Compression-and-Amplification (RTPCA) transformer, an approach that tackles a critical issue in current transformer-based methods: the underutilization of intra-and inter-block relations through attention mechanisms. The cornerstone of our approach is the meticulously designed Temporal Pyramidal Compression-and-Amplification (TPCA) module, which ingeniously leverages a temporal pyramid paradigm to significantly enhance multi-scale key and value representations from intra-block attention. Recognizing that focusing solely on individual modules while overlooking their interconnections can limit performance, we introduce the Cross-Layer Refinement (XLR) module. This carefully crafted component is designed to amplify inter-block communication by linking keys and values across adjacent blocks, creating more coherent attention patterns. The seamless integration of TPCA and XLR results in a powerful synergy that facilitates a rich semantic representation through the dynamic interaction of queries, keys, and values. This synergistic approach enables the RTPCA transformer to achieve remarkable performance on leading benchmarks, such as Human3.6M, HumanEva-I, and MPI-INF-3DHP, with only a small computational overhead. We demonstrate the effectiveness of the RTPCA transformer through extensive experiments and comparisons with state-of-the-art methods. The source code is available at https://github.com/hbing-l/RTPCA.git.

Text-to-Avatar generation has recently made significant strides due to advancements in diffusion models. However, most existing works remain constrained by limited diversity, producing avatars with subtle differences in appearance for a given text prompt. We design DivAvatar, a novel framework that generates diverse avatars, empowering 3D creatives with a multitude of distinct and richly varied 3D avatars from a single text prompt. Different from most existing works that exploit scene-specific 3D representations such as NeRF, DivAvatar finetunes a 3D generative model (i.e., EVA3D), allowing diverse avatar generation from simply noise sampling in inference time. DivAvatar has two key designs that help achieve generation diversity and visual quality. The first is a noise sampling technique during

Generative diffusion prior captured from the off-the-shelf denoising diffusion generative model has recently attained significant interest. However, several attempts have been made to adopt diffusion models to noisy inverse problems either fail to achieve satisfactory results or require a few thousand iterations to achieve high-quality reconstructions. In this work, we propose a diffusion-based image restoration with error contraction and error correction (DiffECC) method. Two strategies are introduced to contract the restoration error in the posterior sampling process. First, we combine existing CNN-based approaches with diffusion models to ensure data consistency from the beginning. Second, to amplify the error contraction effects of the noise, a restart sampling algorithm is designed. In the error correction strategy, the estimation-correction idea is proposed on both the data term and the prior term. Solving them iteratively within the diffusion sampling framework leads to superior image generation results. Experimental results for image restoration tasks such as super-resolution (SR), Gaussian deblurring, and motion deblurring demonstrate that our approach can reconstruct high-quality images compared with state-of-the-art sampling-based diffusion models.

Digital image enhancement aims to deliver visually striking, pleasing images that align with human perception. While global techniques can elevate the image's overall aesthetics, fine-grained color enhancement can further boost visual appeal and expressiveness. However, colorists frequently face challenges in achieving accurate, localized color adjustments. Direct composition of these local edits can result in spatial color inconsistencies. Existing methods, including color style transfer and image harmonization, exhibit inconsistencies, especially at boundary regions. Addressing this, we present ChromaFusionNet (CFNet), a novel approach that views the color fusion problem through the lens of image color inpainting. Built on the Vision Transformer architecture, CFNet captures global context and delivers high-fidelity outputs, seamlessly blending colors while preserving boundary integrity. Empirical studies on ImageNet and COCO datasets demonstrate CFNet's superiority over existing methods in maintaining color harmony and color fidelity. Robustness evaluations and user studies have further validated the effectiveness of CFNet. In conclusion, CFNet introduces an innovative approach to seamless, fine-grained color fusion, paving the way for advancements in the domain of fine-grained color editing. Code and pretrained models are available at our project page: https://yidong.pro/projects/cfnet.

Multimodal Large Language Model (MLLMs) leverages Large Language Models as a cognitive framework for diverse visual-language tasks. Recent efforts have been made to equip MLLMs with visual perceiving and grounding capabilities. However, there still remains a gap in providing fine-grained pixel-level perceptions and extending interactions beyond text-specific inputs. In this work, we propose AnyRef, a general MLLM model that can generate pixel-wise object perceptions and natural language descriptions from multi-modality references, such as texts, boxes, images, or audio. This innovation empowers users with greater flexibility to engage with the model beyond textual and regional prompts, without modality-specific designs. Through our proposed refocusing mechanism, the generated grounding output is guided to better focus on the referenced object, implicitly incorporating additional pixel-level supervision. This simple modification utilizes attention scores generated during the inference of LLM, eliminating the need for extra computations while exhibiting performance enhancements in both grounding masks and referring expressions. With only publicly available training data, our model achieves state-of-the-art results across multiple benchmarks, including diverse modality referring segmentation and region-level referring expression generation. Code and models are available at https://github.com/jwh97nn/AnyRef

Text-guided domain adaptation and generation of 3D-aware portraits find many applications in various fields. However, due to the lack of training data and the challenges in handling the high variety of geometry and appearance, the existing methods for these tasks suffer from issues like inflexibility, instability, and low fidelity. In this paper, we propose a novel framework DiffusionGAN3D, which boosts text-guided 3D domain adaptation and generation by combining 3D GANs and diffusion priors. Specifically, we in-tegrate the pre-trained 3D generative models (e.g., EG3D) and text-to-image diffusion models. The former provides a strong foundation for stable and high-quality avatar generation from text. And the diffusion models in turn offer powerful priors and guide the 3D generator finetuning with informative direction to achieve flexible and efficient text-guided domain adaptation. To enhance the diversity in do-main adaptation and the generation capability in text-to-avatar, we introduce the relative distance loss and case-specific learnable triplane respectively. Besides, we design a progressive texture refinement module to improve the texture quality for both tasks above. Extensive experiments demonstrate that the proposed framework achieves excellent results in both domain adaptation and text-to-avatar tasks, outperforming existing methods in terms of generation quality and efficiency. The project homepage is at https://younglbw.github.io/DiffusionGAN3D-homepage/.

We present En3D, an enhanced generative scheme for sculpting high-quality 3D human avatars. Unlike previous works that rely on scarce 3D datasets or limited 2D collections with imbalanced viewing angles and imprecise pose priors, our approach aims to develop a zero-shot 3D generative scheme capable of producing visually realistic, ge-ometrically accurate and content-wise diverse 3D humans without directly relying on pre-existing 3D or 2D assets. To address this challenge, we introduce a meticulously crafted workflow that implements accurate physical modeling to learn the enhanced 3D generative model from synthetic 2D data. During inference, we integrate optimization modules to bridge the gap between realistic appearances and coarse 3D shapes. Specifically, En3D comprises three modules: a 3D generator that accurately models generalizable 3D humans with realistic appearance from synthesized balanced, diverse, and structured human images; a geometry sculptor that enhances shape quality using multi-view normal constraints for intricate human structure; and a texturing module that disentangles explicit texture maps with fidelity and editability, leveraging semantical UV partitioning and a differentiable rasterizer: Experimental results show that our approach significantly outperforms prior works in terms of image quality, geometry accuracy and content diversity. We also showcase the applicability of our generated avatars for animation and editing, as well as the scalability of our approach for content-style free adaptation.

Visual content creation has aroused a surge of interest given its applications in mobile photography and AR/VR. Portrait style transfer and 3D recovery from monocular images as two representative tasks have so far evolved independently. In this paper, we make a connection between the two, and tackle the challenging task of 3D portrait styl-ization - modeling high-fidelity 3D stylized avatars from captured 2D portrait images. However, naively combining the techniques from the two isolated areas may suf-fer from either inadequate stylization or absence of 3D as-sets. To this end, we propose 3DToonify, a new framework that introduces a progressive training scheme to achieve 3D style adaption on spatial neural representation (SNR). SNR is constructed with implicit fields and they are dynamically optimized by the progressive training scheme, which consists of three stages: guided prior learning, deformable geometry adaption and explicit texture adaption. In this way, stylized geometry and texture are learned in SNR in an explicit and structured way with only a single stylized exemplar needed. Moreover, our method obtains style-adaptive underlying structures (i.e., deformable geometry and exaggerated texture) and view-consistent styl-ized avatar rendering from arbitrary novel viewpoints. Both qualitative and quantitative experiments have been conducted to demonstrate the effectiveness and superiority of our method for automatically generating exemplar-guided 3D stylized avatars.

The transformer model has gained widespread adoption in computer vision tasks in recent times. However, due to the quadratic time and memory complexity of self-attention, which is proportional to the number of input tokens, most existing Vision Transformers (ViTs) encounter challenges in achieving efficient performance in practical industrial deployment scenarios, such as TensorRT and CoreML, where traditional CNNs excel. Although some recent attempts have been made to design CNN-Transformer hybrid architectures to tackle this problem, their overall performance has not met expectations. To tackle these challenges, we propose an efficient hybrid ViT architecture named FMViT. This approach enhances the model’s expressive power by blending high-frequency features and low-frequency features with varying frequencies, enabling it to capture both local and global information effectively. Additionally, we introduce deploy-friendly mechanisms such as Multiple groups of MLP (gMLP) Reparameterization, Lightweight Multi-head Self-Attention (RLMHSA), and Convolutional Fusion Block (CFB) to further improve the model’s performance and reduce computational overhead. Our experiments demonstrate that FMViT surpasses existing CNNs, ViTs, and CNN-Transformer hybrid architectures in terms of latency/accuracy trade-offs for various vision tasks. On the TensorRT platform, FMViT outperforms Resnet101 by 2.5% (83.3% vs. 80.8%) in top-1 accuracy on the ImageNet dataset while maintaining similar inference latency. Moreover, FMViT achieves comparable performance with EfficientNet-B5 but with a 43% improvement in inference speed. On CoreML, FMViT outperforms MobileOne by 2.6% in top-1 accuracy on the ImageNet dataset, with inference latency comparable to MobileOne (78.5% vs. 75.9%).

Data pruning aims to obtain lossless performances with less overall cost. A common approach is to filter out samples that make less contribution to the training. This could lead to gradient expectation bias compared to the original data. To solve this problem, we propose InfoBatch, a novel framework aiming to achieve lossless training acceleration by unbiased dynamic data pruning. Specifically, InfoBatch randomly prunes a portion of less informative samples based on the loss distribution and rescales the gradients of the remaining samples to approximate the original gradient. As a plug-and-play and architecture-agnostic framework, InfoBatch consistently obtains lossless training results on classification, semantic segmentation, vision pertaining, and instruction fine-tuning tasks. On CIFAR10/100, ImageNet- 1K, and ADE20K, InfoBatch losslessly saves 40% overall cost. For pertaining MAE and diffusion model, InfoBatch can respectively save 24.8% and 27% cost. For LLaMA instruction fine-tuning, combining InfoBatch and the recent coreset selection method (DQ) can achieve 10 times acceleration. Our results encourage more exploration on the data efficiency aspect of large model training. Code is publicly available at NUS-HPC-AI-Lab/InfoBatch.

Diffusion model based Text-to-Image has achieved impressive achievements recently. Although current technology for synthesizing images is highly advanced and capable of generating images with high fidelity, it is still possible to give the show away when focusing on the text area in the generated image, as synthesized text often contains blurred, unreadable, or incorrect characters, making visual text generation one of the most challenging issues in this field. To address this issue, we introduce AnyText, a diffusion-based multilingual visual text generation and editing model, that focuses on rendering accurate and coherent text in the image. AnyText comprises a diffusion pipeline with two primary elements: an auxiliary latent module and a text embedding module. The former uses inputs like text glyph, position, and masked image to generate latent features for text generation or editing. The latter employs an OCR model for encoding stroke data as embeddings, which blend with image caption embeddings from the tokenizer to generate texts that seamlessly integrate with the background. We employed text-control diffusion loss and text perceptual loss for training to further enhance writing accuracy. AnyText can write characters in multiple languages, to the best of our knowledge, this is the first work to address multilingual visual text generation. It is worth mentioning that AnyText can be plugged into existing diffusion models from the community for rendering or editing text accurately. After conducting extensive evaluation experiments, our method has outperformed all other approaches by a significant margin. Additionally, we contribute the first large-scale multilingual text images dataset, AnyWord-3M, containing 3 million image-text pairs with OCR annotations in multiple languages. Based on AnyWord-3M dataset, we propose AnyText-benchmark for the evaluation of visual text generation accuracy and quality. Our project will be open-sourced soon to improve and promote the development of text generation technology.

Despite the impressive performance of Multi-view Stereo (MVS) approaches given plenty of training samples, the performance degradation when generalizing to unseen domains has not been clearly explored yet. In this work, we focus on the domain generalization problem in MVS. To evaluate the generalization results, we build a novel MVS domain generalization benchmark including synthetic and real-world datasets. In contrast to conventional domain generalization benchmarks, we consider a more realistic but challenging scenario, where only one source domain is available for training. The MVS problem can be analogized back to the feature matching task, and maintaining robust feature consistency among views is an important factor for improving generalization performance. To address the domain generalization problem in MVS, we propose a novel MVS framework, namely RobustMVS1. A Depth-Clustering-guided Whitening (DCW) loss is further introduced to preserve the feature consistency among different views, which decorrelates multi-view features from viewpoint-specific style information based on geometric priors from depth maps. The experimental results further show that our method achieves superior performance on the domain generalization benchmark2.

Semantic segmentation in driving scenarios is important for modern autonomous driving technology. While the existing methods have shown promising results in segmenting normal-condition images, their performance in adverse scenes remains unsatisfactory due to limited visual field and lack of annotation. To address this issue, we propose an unsupervised domain adaptation semantic segmentation method with the transformer architecture, namely ACSegFormer, for driving-scene adverse conditions, aiming at mining image features in visually restricted scenes. Three effective training strategies are proposed in ACSegFormer to learn the latent image context relations and to reduce the gaps between different domains: an entropy-based pseudo label correction scheme that refines the target domain predictions with the normal reference predictions, an optimal transport-based inter-domain alignment module that performs domain alignment on the outputs of transformer encoder, and a masked context learning module that enhances the model’s ability to perceive the missing information of target domain image. Our ACSegFormer has no additional training parameters on top of the existing transformer segmentation framework, which can be easily used for self-training-based unsupervised domain adaptation approaches. The experimental results show that our ACSegFormer achieves state-of-the-art performance on driving-scene segmentation benchmarks in adverse conditions, including Dark Zurich and ACDC. Codes and models are available athttps://github.com/wenyyu/ACSegFormer.

Generative neural radiance fields (NeRF) bring image generation into the 3D era, which have delivered impressive generation quality and 3D consistency, especially in the face generation domain. Upon pre-trained generative NeRF, 3D-aware image editing has been explored and achieved promising performance via manipulating semantic maps or attributes. However, a more flexible editing interface, text, remains under-explored in the context of 3D-aware image editing. In this work, we leverage the Contrastive Language-Image Pre-training (CLIP) model to achieve 3D-aware image editing in pre-trained generative NeRF models given a target text prompt. To achieve accurate and controllable geometry editing, we propose MorphNeRF, a learnable morphing network that morphs the 3D geometry of images toward the target descriptions via generative NeRF. Different from prior studies that achieve image editing by manipulating latent codes or directly finetuning pre-trained models, morphing the geometry can better preserve the texture of the source image and facilitate the control of editing strength by adjusting the weight of morphing maps explicitly. Extensive experiments and comparisons show that the proposed MorphNeRF achieves superior image editing performance.

Data uncertainty is commonly observed in the images for face recognition (FR). However, deep learning algorithms often make predictions with high confidence even for uncertain or irrelevant inputs. Intuitively, FR algorithms can benefit from both the estimation of uncertainty and the detection of out-of-distribution (OOD) samples. Taking a probabilistic view of the current classification model, the temperature scalar is exactly the scale of uncertainty noise implicitly added in the softmax function. Meanwhile, the uncertainty of images in a dataset should follow a prior distribution. Based on the observation, a unified framework for uncertainty modeling and FR, Random Temperature Scaling (RTS), is proposed to learn a reliable FR algorithm. The benefits of RTS are two-fold. (1) In the training phase, it can adjust the learning strength of clean and noisy samples for stability and accuracy. (2) In the test phase, it can provide a score of confidence to detect uncertain, low-quality and even OOD samples, without training on extra labels. Extensive experiments on FR benchmarks demonstrate that the magnitude of variance in RTS, which serves as an OOD detection metric, is closely related to the uncertainty of the input image. RTS can achieve top performance on both the FR and OOD detection tasks. Moreover, the model trained with RTS can perform robustly on datasets with noise. The proposed module is light-weight and only adds negligible computation cost to the model.

Recent attention in instance segmentation has focused on query-based models. Despite being non-maximum suppression (NMS)-free and end-to-end, the superiority of these models on high-accuracy real-time benchmarks has not been well demonstrated. In this paper, we show the strong potential of query-based models on efficient instance segmentation algorithm designs. We present FastInst, a simple, effective query-based framework for real-time instance segmentation. FastInst can execute at a real-time speed (i.e., 32.5 FPS) while yielding an AP of more than 40 (i.e., 40.5 AP) on COCO test-dev without bells and whistles. Specifically, FastInst follows the meta-architecture of recently introduced Mask2Former. Its key designs include instance activation-guided queries, dual-path update strategy, and ground truth mask-guided learning, which enable us to use lighter pixel decoders, fewer Transformer decoder layers, while achieving better performance. The experiments show that FastInst outperforms most state-of-the-art real-time counterparts, including strong fully convolutional baselines, in both speed and accuracy. Code can be found at https://github.com/junjiehe96/FastInst.

Limited by the nature of the low-dimensional representational capacity of 3DMM, most of the 3DMM-based face reconstruction (FR) methods fail to recover high-frequency facial details, such as wrinkles, dimples, etc. Some attempt to solve the problem by introducing detail maps or nonlinear operations, however, the results are still not vivid. To this end, we in this paper present a novel hierarchical representation network (HRN) to achieve accurate and detailed face reconstruction from a single image. Specifically, we implement the geometry disentanglement and introduce the hierarchical representation to fulfill detailed face modeling. Meanwhile, 3D priors of facial details are incorporated to enhance the accuracy and authenticity of the reconstruction results. We also propose a deretouching module to achieve better decoupling of the geometry and appearance. It is noteworthy that our framework can be extended to a multi-view fashion by considering detail consistency of different views. Extensive experiments on two single-view and two multi-view FR benchmarks demonstrate that our method outperforms the existing methods in both reconstruction accuracy and visual effects. Finally, we introduce a high-quality 3D face dataset FaceHD-100 to boost the research of high-fidelity face reconstruction. The project homepage is at https://younglbw.github.io/HRN-homepage/.

End-to-end pedestrian detection focuses on training a pedestrian detection model via discarding the Non-Maximum Suppression (NMS) post-processing. Though a few methods have been explored, most of them still suffer from longer training time and more complex deployment, which cannot be deployed in the actual industrial applications. In this paper, we intend to bridge this gap and propose an Optimal Proposal Learning (OPL) framework for deployable end-to-end pedestrian detection. Specifically, we achieve this goal by using CNN-based light detector and introducing two novel modules, including a Coarse-to-Fine (C2F) learning strategy for proposing precise positive proposals for the Ground-Truth (GT) instances by reducing the ambiguity of sample assignment/output in training/testing respectively, and a Completed Proposal Network (CPN) for producing extra information compensation to further recall the hard pedestrian samples. Extensive experiments are conducted on CrowdHuman, TJU-Ped and Caltech, and the results show that our proposed OPL method significantly outperforms the competing methods.

Existing Visual Object Tracking (VOT) only takes the target area in the first frame as a template. This causes tracking to inevitably fail in fast-changing and crowded scenes, as it cannot account for changes in object appearance between frames. To this end, we revamped the tracking framework with Progressive Context Encoding Transformer Tracker (ProContEXT), which coherently exploits spatial and temporal contexts to predict object motion trajectories. Specifically, ProContEXT leverages a context-aware self-attention module to encode the spatial and temporal context, refining and updating the multi-scale static and dynamic templates to progressively perform accurately tracking. It explores the complementary between spatial and temporal context, raising a new pathway to multi-context modeling for transformer-based trackers. In addition, ProContEXT revised the token pruning technique to reduce computational complexity. Extensive experiments on popular benchmark datasets such as GOT-10k and TrackingNet demonstrate that the proposed ProContEXT achieves state-of-the-art performance1.

Streaming perception is a fundamental task in autonomous driving that requires a careful balance between the latency and accuracy of the autopilot system. However, current methods for streaming perception are limited as they rely only on the current and adjacent two frames to learn movement patterns, which restricts their ability to model complex scenes, often leading to poor detection results. To address this limitation, we propose LongShortNet, a novel dual-path network that captures long-term temporal motion and integrates it with short-term spatial semantics for real-time perception. Our proposed LongShortNet is notable as it is the first work to extend long-term temporal modeling to streaming perception, enabling spatiotemporal feature fusion. We evaluate LongShortNet on the challenging Argoverse-HD dataset and demonstrate that it outperforms existing state-of-the-art methods with almost no additional computational cost.1

Image colorization is a challenging problem due to multi-modal uncertainty and high ill-posedness. Directly training a deep neural network usually leads to incorrect semantic colors and low color richness. While transformer-based methods can deliver better results, they often rely on manually designed priors, suffer from poor generalization ability, and introduce color bleeding effects. To address these issues, we propose DDColor, an end-to-end method with dual decoders for image colorization. Our approach includes a pixel decoder and a query-based color decoder. The former restores the spatial resolution of the image, while the latter utilizes rich visual features to refine color queries, thus avoiding hand-crafted priors. Our two decoders work together to establish correlations between color and multi-scale semantic representations via cross-attention, significantly alleviating the color bleeding effect. Additionally, a simple yet effective colorfulness loss is introduced to enhance the color richness. Extensive experiments demonstrate that DDColor achieves superior performance to existing state-of-the-art works both quantitatively and qualitatively. The codes and models are publicly available at https://github.com/piddnad/DDColor.

Semantic segmentation of point clouds usually requires exhausting efforts of human annotations, hence it attracts wide attention to the challenging topic of learning from unlabeled or weaker forms of annotations. In this paper, we take the first attempt for fully unsupervised semantic segmentation of point clouds, which aims to delineate semantically meaningful objects without any form of annotations. Previous works of unsupervised pipeline on 2D images fails in this task of point clouds, due to: 1) Clustering Ambiguity caused by limited magnitude of data and imbalanced class distribution; 2) Irregularity Ambiguity caused by the irregular sparsity of point cloud. Therefore, we propose a novel framework, PointDC, which is comprised of two steps that handle the aforementioned problems respectively: Cross-Modal Distillation (CMD) and Super-Voxel Clustering (SVC). In the first stage of CMD, multi-view visual features are back-projected to the 3D space and aggregated to a unified point feature to distill the training of the point representation. In the second stage of SVC, the point features are aggregated to super-voxels and then fed to the iterative clustering process for excavating semantic classes. PointDC1yields a significant improvement over the prior state-of-the-art unsupervised methods, on both the ScanNet-v2 (+18.4 mIoU) and S3DIS (+11.5 mIoU) semantic segmentation benchmarks.

Vision Transformers (ViTs) have demonstrated powerful representation ability in various visual tasks thanks to their intrinsic data-hungry nature. However, we unexpectedly find that ViTs perform vulnerably when applied to face recognition (FR) scenarios with extremely large datasets. We investigate the reasons for this phenomenon and discover that the existing data augmentation approach and hard sample mining strategy are incompatible with ViTs-based FR backbone due to the lack of tailored consideration on preserving face structural information and leveraging each local token information. To remedy these problems, this paper proposes a superior FR model called TransFace, which employs a patch-level data augmentation strategy named DPAP and a hard sample mining strategy named EHSM. Specially, DPAP randomly perturbs the amplitude information of dominant patches to expand sample diversity, which effectively alleviates the overfitting problem in ViTs. EHSM utilizes the information entropy in the local tokens to dynamically adjust the importance weight of easy and hard samples during training, leading to a more stable prediction. Experiments on several benchmarks demonstrate the superiority of our TransFace. Code and models are available at https://github.com/DanJun6737/TransFace.

Depth-aware panoptic segmentation is an emerging topic in computer vision which combines semantic and geometric understanding for more robust scene interpretation. Recent works pursue unified frameworks to tackle this challenge but mostly still treat it as two individual learning tasks, which limits their potential for exploring cross-domain information. We propose a deeply unified framework for depth-aware panoptic segmentation, which performs joint segmentation and depth estimation both in a persegment manner with identical object queries. To narrow the gap between the two tasks, we further design a geometric query enhancement method, which is able to integrate scene geometry into object queries using latent representations. In addition, we propose a bi-directional guidance learning approach to facilitate cross-task feature learning by taking advantage of their mutual relations. Our method sets the new state of the art for depth-aware panoptic segmentation on both Cityscapes-DVPS and SemKITTI-DVPS datasets. Moreover, our guidance learning approach is shown to deliver performance improvement even under incomplete supervision labels. Code and models are available at https://github.com/jwh97nn/DeepDPS.

Retouching images is an essential aspect of enhancing the visual appeal of photos. Although users often share common aesthetic preferences, their retouching methods may vary based on their individual preferences. Therefore, there is a need for white-box approaches that produce satisfying results and enable users to conveniently edit their images simultaneously. Recent white-box retouching methods rely on cascaded global filters that provide image-level filter arguments but cannot perform fine-grained retouching. In contrast, colorists typically employ a divide-andconquer approach, performing a series of region-specific fine-grained enhancements when using traditional tools like Davinci Resolve. We draw on this insight to develop a white-box framework for photo retouching using parallel region-specific filters, called RSFNet. Our model generates filter arguments (e.g., saturation, contrast, hue) and attention maps of regions for each filter simultaneously. Instead of cascading filters, RSFNet employs linear summations of filters, allowing for a more diverse range of filter classes that can be trained more easily. Our experiments demonstrate that RSFNet achieves state-of-the-art results, offering satisfying aesthetic appeal and increased user convenience for editable white-box retouching. Code is available at https://github.com/Vicky0522/RSFNet.

Unsupervised domain adaptation (UDA) has proven to be highly effective in transferring knowledge from a label-rich source domain to a label-scarce target domain. However, the presence of additional novel categories in the target domain has led to the development of open-set domain adaptation (ODA) and universal domain adaptation (UNDA). Existing ODA and UNDA methods treat all novel categories as a single, unified unknown class and attempt to detect it during training. However, we found that domain variance can lead to more significant view-noise in unsupervised data augmentation, which affects the effectiveness of contrastive learning (CL) and causes the model to be overconfident in novel category discovery. To address these issues, a framework named Soft-contrastive All-in-one Network (SAN) is proposed for ODA and UNDA tasks. SAN includes a novel data-augmentation-based soft contrastive learning (SCL) loss to fine-tune the backbone for feature transfer and a more human-intuitive classifier to improve new class discovery capability. The SCL loss weakens the adverse effects of the data augmentation view-noise problem which is amplified in domain transfer tasks. The All-in-One (AIO) classifier overcomes the overconfidence problem of current mainstream closed-set and open-set classifiers. Visualization and ablation experiments demonstrate the effectiveness of the proposed innovations. Furthermore, extensive experiment results on ODA and UNDA show that SAN outperforms existing state-of-the-art methods.

Human pose estimation is a challenging task due to its structured data sequence nature. Existing methods primarily focus on pair-wise interaction of body joints, which is insufficient for scenarios involving overlapping joints and rapidly changing poses. To overcome these issues, we introduce a novel approach, the High-order Directed Transformer (HDFormer), which leverages high-order bone and joint relationships for improved pose estimation. Specifically, HDFormer incorporates both self-attention and high-order attention to formulate a multi-order attention module. This module facilitates first-order "joint-joint", second-order "bone-joint", and high-order "hyperbone-joint" interactions, effectively addressing issues in complex and occlusion-heavy situations. In addition, modern CNN techniques are integrated into the transformer-based architecture, balancing the trade-off between performance and efficiency. HDFormer significantly outperforms state-of-the-art (SOTA) models on Human3.6M and MPI-INF-3DHP datasets, requiring only 1/10 of the parameters and significantly lower computational costs. Moreover, HDFormer demonstrates broad real-world applicability, enabling real-time, accurate 3D pose estimation. The source code is in https://github.com/hyer/HDFormer.

The core of Multi-view Stereo(MVS) is the matching process among reference and source pixels. Cost aggregation plays a significant role in this process, while previous methods focus on handling it via CNNs. This may inherit the natural limitation of CNNs that fail to discriminate repetitive or incorrect matches due to limited local receptive fields. To handle the issue, we aim to involve Transformer into cost aggregation. However, another problem may occur due to the quadratically growing computational complexity caused by Transformer, resulting in memory overflow and inference latency. In this paper, we overcome these limits with an efficient Transformer-based cost aggregation network, namely CostFormer. The Residual Depth-Aware Cost Transformer(RDACT) is proposed to aggregate long-range features on cost volume via self-attention mechanisms along the depth and spatial dimensions. Furthermore, Residual Regression Transformer(RRT) is proposed to enhance spatial attention. The proposed method is a universal plug-in to improve learning-based MVS methods.

In the realm of autonomous driving, real-time perception or streaming perception remains under-explored. This research introduces DAMO-StreamNet, a novel framework that merges the cutting-edge elements of the YOLO series with a detailed examination of spatial and temporal perception techniques. DAMO-StreamNet's main inventions include: (1) a robust neck structure employing deformable convolution, bolstering receptive field and feature alignment capabilities; (2) a dual-branch structure synthesizing short-path semantic features and long-path temporal features, enhancing the accuracy of motion state prediction; (3) logits-level distillation facilitating efficient optimization, which aligns the logits of teacher and student networks in semantic space; and (4) a real-time prediction mechanism that updates the features of support frames with the current frame, providing smooth streaming perception during inference. Our testing shows that DAMO-StreamNet surpasses current state-of-the-art methodologies, achieving 37.8% (normal size (600, 960)) and 43.3% (large size (1200, 1920)) sAP without requiring additional data. This study not only establishes a new standard for real-time perception but also offers valuable insights for future research. The source code is at https://github.com/zhiqic/DAMO-StreamNet.

In the realm of facial analysis, accurate landmark detection is crucial for various applications, ranging from face recognition and expression analysis to animation. Conventional heatmap or coordinate regression-based techniques, however, often face challenges in terms of computational burden and quantization errors. To address these issues, we present the KeyPoint Positioning System (KeyPosS) - a groundbreaking facial landmark detection framework that stands out from existing methods. The framework utilizes a fully convolutional network to predict a distance map, which computes the distance between a Point of Interest (POI) and multiple anchor points. These anchor points are ingeniously harnessed to triangulate the POI's position through the True-range Multilateration algorithm. Notably, the plug-and-play nature of KeyPosS enables seamless integration into any decoding stage, ensuring a versatile and adaptable solution. We conducted a thorough evaluation of KeyPosS's performance by benchmarking it against state-of-the-art models on four different datasets. The results show that KeyPosS substantially outperforms leading methods in low-resolution settings while requiring a minimal time overhead.1 The code is available at https://github.com/zhiqic/KeyPosS.

The current 3D human pose estimators face challenges in adapting to new datasets due to the scarcity of 2D-3D pose pairs in target domain training sets. We present the Multi-Hypothesis Pose Synthesis Domain Adaptation (PoSynDA) framework to overcome this issue without extensive target domain annotation. Utilizing a diffusion-centric structure, PoSynDA simulates the 3D pose distribution in the target domain, filling the data diversity gap. By incorporating a multi-hypothesis network, it creates diverse pose hypotheses and aligns them with the target domain. Target-specific source augmentation obtains the target domain distribution data from the source domain by decoupling the scale and position parameters. The teacher-student paradigm and low-rank adaptation further refine the process. PoSynDA demonstrates competitive performance on benchmarks, such as Human3.6M, MPI-INF-3DHP, and 3DPW, even comparable with the target-trained MixSTE model. This work paves the way for the practical application of 3D human pose estimation1. The source code is available at https://github.com/hbing-l/PoSynDA.

Significant progress has been witnessed in learning-based Multi-view Stereo (MVS) under supervised and unsupervised settings. To combine their respective merits in accuracy and completeness, meantime reducing the demand for expensive labeled data, this paper explores the problem of learning-based MVS in a semi-supervised setting that only a tiny part of the MVS data is attached with dense depth ground truth. However, due to huge variation of scenarios and flexible settings in views, it may break the basic assumption in classic semi-supervised learning, that unlabeled data and labeled data share the same label space and data distribution, named as semi-supervised distribution-gap ambiguity in the MVS problem. To handle these issues, we propose a novel semi-supervised distribution-augmented MVS framework, namely SDA-MVS. For the simple case that the basic assumption works in MVS data, consistency regularization encourages the model predictions to be consistent between original sample and randomly augmented sample. For further troublesome case that the basic assumption is conflicted in MVS data, we propose a novel style consistency loss to alleviate the negative effect caused by the distribution gap. The visual style of unlabeled sample is transferred to labeled sample to shrink the gap, and the model prediction of generated sample is further supervised with the label in original labeled sample. The experimental results in semi-supervised settings of multiple MVS datasets show the superior performance of the proposed method. With the same settings in backbone network, our proposed SDA-MVS outperforms its fully-supervised and unsupervised baselines.

Semantic segmentation on driving-scene images is vital for autonomous driving. Although encouraging performance has been achieved on daytime images, the performance on nighttime images are less satisfactory due to the insufficient exposure and the lack of labeled data. To address these issues, we present an add-on module called dual image-adaptive learnable filters (DIAL-Filters) to improve the semantic segmentation in nighttime driving conditions, aiming at exploiting the intrinsic features of driving-scene images under different illuminations. DIAL-Filters consist of two parts, including an image-adaptive processing module (IAPM) and a learnable guided filter (LGF). With DIAL-Filters, we design both unsupervised and supervised frameworks for nighttime driving-scene segmentation, which can be trained in an end-to-end manner. Specifically, the IAPM module consists of a small convolutional neural network with a set of differentiable image filters, where each image can be adaptively enhanced for better segmentation with respect to the different illuminations. The LGF is employed to enhance the output of segmentation network to get the final segmentation result. The DIAL-Filters are light-weight and efficient and they can be readily applied for both daytime and nighttime images. Our experiments show that DAIL-Filters can significantly improve the supervised segmentation performance on ACDC_Night and NightCity datasets, while it demonstrates the state-of-the-art performance on unsupervised nighttime semantic segmentation on Dark Zurich and Nighttime Driving testbeds. Codes and models are available athttps://github.com/wenyyu/IA-Seg.

This paper proposes a novel active boundary loss for semantic segmentation. It can progressively encourage the alignment between predicted boundaries and ground-truth boundaries during end-to-end training, which is not explicitly enforced in commonly used cross-entropy loss. Based on the predicted boundaries detected from the segmentation results using current network parameters, we formulate the boundary alignment problem as a differentiable direction vector prediction problem to guide the movement of predicted boundaries in each iteration. Our loss is model-agnostic and can be plugged in to the training of segmentation networks to improve the boundary details. Experimental results show that training with the active boundary loss can effectively improve the boundary F-score and mean Intersection-over-Union on challenging image and video object segmentation datasets.

Photo retouching finds many applications in various fields. However, most existing methods are designed for global retouching and seldom pay attention to the local region, while the latter is actually much more tedious and time-consuming in photography pipelines. In this paper, we propose a novel adaptive blend pyramid network, which aims to achieve fast local retouching on ultra high-resolution photos. The network is mainly composed of two components: a context-aware local retouching layer (LRL) and an adaptive blend pyramid layer (BPL). The LRL is designed to implement local retouching on low-resolution images, giving full consideration of the global context and local texture information, and the BPL is then developed to progressively expand the low-resolution results to the higher ones, with the help of the proposed adaptive blend module and refining module. Our method outperforms the existing methods by a large margin on two local photo retouching tasks and exhibits excellent performance in terms of running speed, achieving real-time inference on 4K images with a single NVIDIA Tesla P100 GPU. Moreover, we introduce the first high-definition cloth retouching dataset CRHD-3K to promote the research on local photo retouching. The dataset is available at https://github.com/youngLbw/crhd-3K.

In this paper, we present a general-purpose solution to cartoon image synthesis with unpaired training data. In contrast to previous works learning pre-defined cartoon styles for specified usage scenarios (portrait or scene), we aim to train a common cartoon translator which can not only simultaneously render exaggerated anime faces and realistic cartoon scenes, but also provide flexible user controls for desired cartoon styles. It is challenging due to the complexity of the task and the absence of paired data. The core idea of the proposed method is to introduce gated cycle mapping, that utilizes a novel gated mapping unit to produce the category-specific style code and embeds this code into cycle networks to control the translation process. For the concept of category, we classify images into different categories (e.g., 4 types: photo/cartoon portrait/scene) and learn finer-grained category translations rather than overall mappings between two domains (e.g., photo and cartoon). Furthermore, the proposed method can be easily extended to cartoon video generation with an auxiliary dataset and a new adaptive style loss. Experimental results demonstrate the superiority of the proposed method over the state of the art and validate its effectiveness in the brand-new task of general cartoon image synthesis.

Body reshaping is an important procedure in portrait photo retouching. Due to the complicated structure and multifarious appearance of human bodies, existing methods either fall back on the 3D domain via body morphable model or resort to keypoint-based image deformation, leading to inefficiency and unsatisfied visual quality. In this paper, we address these limitations by formulating an end-to-end flow generation architecture under the guidance of body structural priors, including skeletons and Part Affinity Fields, and achieve unprecedentedly controllable performance under arbitrary poses and garments. A compositional attention mechanism is introduced for capturing both visual perceptual correlations and structural associations of the human body to reinforce the manipulation consistency among related parts. For a comprehensive evaluation, we construct the first large-scale body reshaping dataset, namely BR-5K, which contains 5,000 portrait photos as well as professionally retouched targets. Extensive experiments demonstrate that our approach significantly outperforms existing state-of-the-art methods in terms of visual performance, controllability, and efficiency. The dataset is available at our website: https://github.com/JianqiangRen/FlowBasedBodyReshaping.

Leveraging StyleGAN's expressivity and its disentangled latent codes, existing methods can achieve realistic editing of different visual attributes such as age and gender of facial images. An intriguing yet challenging problem arises: Can generative models achieve counterfactual editing against their learnt priors? Due to the lack of counterfactual samples in natural datasets, we investigate this problem in a text-driven manner with Contrastive-Language-Image-Pretraining (CLIP), which can offer rich semantic knowledge even for various counterfactual concepts. Different from in-domain manipulation, counterfactual manipulation requires more comprehensive exploitation of semantic knowledge encapsulated in CLIP as well as more delicate handling of editing directions for avoiding being stuck in local minimum or undesired editing. To this end, we design a novel contrastive loss that exploits predefined CLIP-space directions to guide the editing toward desired directions from different perspectives. In addition, we design a simple yet effective scheme that explicitly maps CLIP embeddings (of target text) to the latent space and fuses them with latent codes for effective latent code optimization and accurate editing. Extensive experiments show that our design achieves accurate and realistic editing while driving by target texts with various counterfactual concepts.

Inferring the scene illumination from a single image is an essential yet challenging task in computer vision and computer graphics. Existing works estimate lighting by regressing representative illumination parameters or generating illumination maps directly. However, these methods often suffer from poor accuracy and generalization. This paper presents Geometric Mover's Light (GMLight), a lighting estimation framework that employs a regression network and a generative projector for effective illumination estimation. We parameterize illumination scenes in terms of the geometric light distribution, light intensity, ambient term, and auxiliary depth, which can be estimated by a regression network. Inspired by the earth mover's distance, we design a novel geometric mover's loss to guide the accurate regression of light distribution parameters. With the estimated light parameters, the generative projector synthesizes panoramic illumination maps with realistic appearance and high-frequency details. Extensive experiments show that GMLight achieves accurate illumination estimation and superior fidelity in relighting for 3D object insertion. The codes are available at https://github.com/fnzhan/Illumination-Estimation.

This paper introduces DCT-Net, a novel image translation architecture for few-shot portrait stylization. Given limited style exemplars (~100), the new architecture can produce high-quality style transfer results with advanced ability to synthesize high-fidelity contents and strong generality to handle complicated scenes (e.g., occlusions and accessories). Moreover, it enables full-body image translation via one elegant evaluation network trained by partial observations (i.e., stylized heads). Few-shot learning based style transfer is challenging since the learned model can easily become overfitted in the target domain, due to the biased distribution formed by only a few training examples. This paper aims to handle the challenge by adopting the key idea of "calibration first, translation later" and exploring the augmented global structure with locally-focused translation. Specifically, the proposed DCT-Net consists of three modules: a content adapter borrowing the powerful prior from source photos to calibrate the content distribution of target samples; a geometry expansion module using affine transformations to release spatially semantic constraints; and a texture translation module leveraging samples produced by the calibrated distribution to learn a fine-grained conversion. Experimental results demonstrate the proposed method's superiority over the state of the art in head stylization and its effectiveness on full image translation with adaptive deformations. Our code is publicly available at https://github.com/menyifang/DCT-Net.

Illumination estimation from a single image is critical in 3D rendering and it has been investigated extensively in the computer vision and computer graphic research community. On the other hand, existing works estimate illumination by either regressing light parameters or generating illumination maps that are often hard to optimize or tend to produce inaccurate predictions. We propose Earth Mover’s Light (EMLight), an illumination estimation framework that leverages a regression network and a neural projector for accurate illumination estimation. We decompose the illumination map into spherical light distribution, light intensity and the ambient term, and define the illumination estimation as a parameter regression task for the three illumination components. Motivated by the Earth Mover's distance, we design a novel spherical mover's loss that guides to regress light distribution parameters accurately by taking advantage of the subtleties of spherical distribution. Under the guidance of the predicted spherical distribution, light intensity and ambient term, the neural projector synthesizes panoramic illumination maps with realistic light frequency. Extensive experiments show that EMLight achieves accurate illumination estimation and the generated relighting in 3D object embedding exhibits superior plausibility and fidelity as compared with state-of-the-art methods.

The existence of noisy labels in real-world data negatively impacts the performance of deep learning models. Although much research effort has been devoted to improving robustness to noisy labels in classification tasks, the problem of noisy labels in deep metric learning (DML) remains open. In this paper, we propose a noise-resistant training technique for DML, which we name Probabilistic Ranking-based Instance Selection with Memory (PRISM). PRISM identifies noisy data in a minibatch using average similarity against image features extracted by several previous versions of the neural network. These features are stored in and retrieved from a memory bank. To alleviate the high computational cost brought by the memory bank, we introduce an acceleration method that replaces individual data points with the class centers. In extensive comparisons with 12 existing approaches under both synthetic and real-world label noise, PRISM demonstrates superior performance of up to 6.06% in Precision@1.

Different from general photo retouching tasks, portrait photo retouching (PPR), which aims to enhance the visual quality of a collection of flat-looking portrait photos, has its special and practical requirements such as human-region priority (HRP) and group-level consistency (GLC). HRP requires that more attention should be paid to human regions, while GLC requires that a group of portrait photos should be retouched to a consistent tone. Models trained on existing general photo retouching datasets, however, can hardly meet these requirements of PPR. To facilitate the research on this high-frequency task, we construct a largescale PPR dataset, namely PPR10K, which is the first of its kind to our best knowledge. PPR10K contains 1, 681 groups and 11, 161 high-quality raw portrait photos in total. High-resolution segmentation masks of human regions are provided. Each raw photo is retouched by three experts, while they elaborately adjust each group of photos to have consistent tones. We define a set of objective measures to evaluate the performance of PPR and propose strategies to learn PPR models with good HRP and GLC performance. The constructed PPR10K dataset provides a good bench-mark for studying automatic PPR methods, and experiments demonstrate that the proposed learning strategies are effective to improve the retouching performance. Datasets and codes are available: https://github.com/csjliang/PPR10K.

Blind face restoration (BFR) from severely degraded face images in the wild is a very challenging problem. Due to the high illness of the problem and the complex unknown degradation, directly training a deep neural network (DNN) usually cannot lead to acceptable results. Existing generative adversarial network (GAN) based methods can produce better results but tend to generate over-smoothed restorations. In this work, we propose a new method by first learning a GAN for high-quality face image generation and embedding it into a U-shaped DNN as a prior decoder, then fine-tuning the GAN prior embedded DNN with a set of synthesized low-quality face images. The GAN blocks are designed to ensure that the latent code and noise input to the GAN can be respectively generated from the deep and shallow features of the DNN, controlling the global face structure, local face details and background of the reconstructed image. The proposed GAN prior embedded network (GPEN) is easy-to-implement, and it can generate visually photo-realistic results. Our experiments demonstrated that the proposed GPEN achieves significantly superior results to state-of-the-art BFR methods both quantitatively and qualitatively, especially for the restoration of severely degraded face images in the wild. The source code and models can be found at https://github.com/yangxy/GPEN.

Despite the great success of GANs in images translation with different conditioned inputs such as semantic segmentation and edge maps, generating high-fidelity realistic images with reference styles remains a grand challenge in conditional image-to-image translation. This paper presents a general image translation framework that incorporates optimal transport for feature alignment between conditional inputs and style exemplars in image translation. The introduction of optimal transport mitigates the constraint of many-to-one feature matching significantly while building up accurate semantic correspondences between conditional inputs and exemplars. We design a novel unbalanced optimal transport to address the transport between features with deviational distributions which exists widely between conditional inputs and exemplars. In addition, we design a semantic-activation normalization scheme that injects style features of exemplars into the image translation process successfully. Extensive experiments over multiple image translation tasks show that our method achieves superior image translation qualitatively and quantitatively as compared with the state-of-the-art.

Image inpainting aims to complete the missing or corrupted regions of images with realistic contents. The prevalent approaches adopt a hybrid objective of reconstruction and perceptual quality by using generative adversarial networks. However, the reconstruction loss and adversarial loss focus on synthesizing contents of different frequencies and simply applying them together often leads to inter-frequency conflicts and compromised inpainting. This paper presents WaveFill, a wavelet-based inpainting network that decomposes images into multiple frequency bands and fills the missing regions in each frequency band separately and explicitly. WaveFill decomposes images by using discrete wavelet transform (DWT) that preserves spatial information naturally. It applies L1 reconstruction loss to the decomposed low-frequency bands and adversarial loss to high-frequency bands, hence effectively mitigate inter-frequency conflicts while completing images in spatial domain. To address the inpainting inconsistency in different frequency bands and fuse features with distinct statistics, we design a novel normalization scheme that aligns and fuses the multi-frequency features effectively. Extensive experiments over multiple datasets show that WaveFill achieves superior image inpainting qualitatively and quantitatively.

Accurate lighting estimation is challenging yet critical to many computer vision and computer graphics tasks such as high-dynamic-range (HDR) relighting. Existing approaches model lighting in either frequency domain or spatial domain which is insufficient to represent the complex lighting conditions in scenes and tends to produce inaccurate estimation. This paper presents NeedleLight, a new lighting estimation model that represents illumination with needlets and allows lighting estimation in both frequency domain and spatial domain jointly. An optimal thresholding function is designed to achieve sparse needlets which trims redundant lighting parameters and demonstrates superior localization properties for illumination representation. In addition, a novel spherical transport loss is designed based on optimal transport theory which guides to regress lighting representation parameters with consideration of the spatial information. Furthermore, we propose a new metric that is concise yet effective by directly evaluating the estimated illumination maps rather than rendered images. Extensive experiments show that NeedleLight achieves superior lighting estimation consistently across multiple evaluation metrics as compared with state-of-the-art methods.

Visual storyline generation is the problem of selecting and sequencing a set of visual materials (i.e. images and video clips) to produce a video to elicit certain cognitive or emotional responses from viewers. In this paper, we enhance the viewing experience of generated visual storylines with the Shot Composition, Selection and Plotting (ShotCSP) approach. Designed for generating promotional videos in ecommerce settings, ShotCSP considers three key film-making principles into the visual storyline generation pipeline: a) proximity-aware scene transition, b) sound logic flow, and c) graphic discontinuity. We propose two novel metrics to enhance viewing experience: 1) Semantic Distance, which measures how related a shot is to the product being promoted; and 2) Salient Region Ratio, which estimates attention to product details in a shot. Through large-scale user evaluation involving 1,748 pairwise comparisons against five state-of-the-art approaches, ShotCSP achieves significantly improved viewing experience. It is a promising approach to enable AI generated promotional videos to benefit e-commerce businesses.

Image inpainting is an underdetermined inverse problem, which naturally allows diverse contents to fill up the missing or corrupted regions realistically. Prevalent approaches using convolutional neural networks (CNNs) can synthesize visually pleasant contents, but CNNs suffer from limited perception fields for capturing global features. With image-level attention, transformers enable to model long-range dependencies and generate diverse contents with autoregressive modeling of pixel-sequence distributions. However, the unidirectional attention in autoregressive transformers is suboptimal as corrupted image regions may have arbitrary shapes with contexts from any direction. We propose BAT-Fill, an innovative image inpainting framework that introduces a novel bidirectional autoregressive transformer (BAT) for image inpainting. BAT utilizes the transformers to learn autoregressive distributions, which naturally allows the diverse generation of missing contents. In addition, it incorporates the masked language model like BERT, which enables bidirectionally modeling of contextual information of missing regions for better image completion. Extensive experiments over multiple datasets show that BAT-Fill achieves superior diversity and fidelity in image inpainting qualitatively and quantitatively.

This paper proposes a novel deep learning-based video object matting method that can achieve temporally coherent matting results. Its key component is an attention-based temporal aggregation module that maximizes image matting networks' strength for video matting networks. This module computes temporal correlations for pixels adjacent to each other along the time axis in feature space, which is robust against motion noises. We also design a novel loss term to train the attention weights, which drastically boosts the video matting performance. Besides, we show how to effectively solve the trimap generation problem by fine-tuning a state-of-the-art video object segmentation network with a sparse set of user-annotated keyframes. To facilitate video matting and trimap generation networks' training, we construct a large-scale video matting dataset with 80 training and 28 validation foreground video clips with ground-truth alpha mattes. Experimental results show that our method can generate high-quality alpha mattes for various videos featuring appearance change, occlusion, and fast motion. Our code and dataset can be found at: https://github.com/yunkezhang/TCVOM

There is an increasing demand for interior design and decorating. The main challenges are where to put the objects and how to put them plausibly in the given domain. In this article, we propose an automatic method for decorating the planes in a given image. We call it Decoration In (DecorIn for short). Given an image, we first extract planes as decorating candidates according to the estimated geometric features. Then we parameterize the planes with an orthogonal and semantically consistent grid. Finally, we compute the position for the decoration, i.e., a decoration box, on the plane by an example-based decorating method which can describe the partial image and compute the similarity between partial scenes. We have conducted comprehensive evaluations and demonstrate our method on a number of applications. Our method is more efficient both in time and economic than generating a layout from scratch.

There is an emerging trend for sellers to use videos to promote their products on e-commerce platforms such as Taobao.com. Current video production workflow includes the production of visual storyline by human directors. We propose a system to automatically generate visual storyline based on the input set of visual materials (e.g. video clips or still images) and then produce a promotional video. In particular, we propose an algorithm called Shot Composition, Selection and Plotting (ShotCSP), which generates visual storylines leveraging film-making principles to improve viewing experience and perceived persuasiveness.

Semantic human matting aims to estimate the per-pixel opacity of the foreground human regions. It is quite challenging that usually requires user interactive trimaps and plenty of high quality annotated data. Annotating such kind of data is labor intensive and requires great skills beyond normal users, especially considering the very detailed hair part of humans. In contrast, coarse annotated human dataset is much easier to acquire and collect from the public dataset. In this paper, we propose to leverage coarse annotated data coupled with fine annotated data to boost end-to-end semantic human matting without trimaps as extra input. Specifically, We train a mask prediction network to estimate the coarse semantic mask using the hybrid data, and then propose a quality unification network to unify the quality of the previous coarse mask outputs. A matting refinement network takes the unified mask and the input image to predict the final alpha matte. The collected coarse annotated dataset enriches our dataset significantly, allows generating high quality alpha matte for real images. Experimental results show that the proposed method performs comparably against state-of-the-art methods. Moreover, the proposed method can be used for refining coarse annotated public dataset, as well as semantic segmentation methods, which reduces the cost of annotating high quality human data to a great extent.

The ability to produce convincing textural details is essential for the fidelity of synthesized person images. Existing methods typically follow a “warping-based” strategy that propagates appearance features through the same pathway used for pose transfer. However, most fine-grained features would be lost during down-sampling, leading to over-smoothed clothes and missing details in the output images. In this paper we presents RATE-Net, a novel framework for synthesizing person images with sharp texture details. The proposed framework leverages an additional texture enhancing module to extract appearance information from the source image and estimate a fine-grained residual texture map, which helps to refine the coarse estimation from the pose transfer module. In addition, we design an effective alternate updating strategy to promote mutual guidance between two modules for better shape and appearance consistency. Experiments conducted on DeepFashion benchmark dataset have demonstrated the superiority of our framework compared with existing networks.

Video editing is currently a highly skill- and time-intensive process. One of the most important tasks in video editing is to compose the visual storyline. This paper outlines Visual Storyline Generator (VSG), an artificial intelligence (AI)-empowered system that automatically generates visual storylines based on a set of images and video footages provided by the user. It is designed to produce engaging and persuasive promotional videos with an easy-to-use interface. In addition, users can be involved in refining the AI-generated visual storylines. The editing results can be used as training data to further improve the AI algorithms in VSG.

Video contents have become a critical tool for promoting products in E-commerce. However, the lack of automatic promotional video generation solutions makes large-scale video-based promotion campaigns infeasible. The first step of automatically producing promotional videos is to generate visual storylines, which is to select the building block footage and place them in an appropriate order. This task is related to the subjective viewing experience. It is hitherto performed by human experts and thus, hard to scale. To address this problem, we propose WundtBackpack, an algorithmic approach to generate storylines based on available visual materials, which can be video clips or images. It consists of two main parts, 1) the Learnable Wundt Curve to evaluate the perceived persuasiveness based on the stimulus intensity of a sequence of visual materials, which only requires a small volume of data to train; and 2) a clustering-based backpacking algorithm to generate persuasive sequences of visual materials while considering video length constraints. In this way, the proposed approach provides a dynamic structure to empower artificial intelligence (AI) to organize video footage in order to construct a sequence of visual stimuli with persuasive power. Extensive real-world experiments show that our approach achieves close to 10% higher perceived persuasiveness scores by human testers, and 12.5% higher expected revenue compared to the best performing state-of-the-art approach.

Neural style transfer has drawn considerable attention from both academic and industrial field. Although visual effect and efficiency have been significantly improved, existing methods are unable to coordinate spatial distribution of visual attention between the content image and stylized image, or render diverse level of detail via different brush strokes. In this paper, we tackle these limitations by developing an attention-aware multi-stroke style transfer model. We first propose to assemble self-attention mechanism into a style-agnostic reconstruction autoencoder framework, from which the attention map of a content image can be derived. By performing multi-scale style swap on content features and style features, we produce multiple feature maps reflecting different stroke patterns. A flexible fusion strategy is further presented to incorporate the salient characteristics from the attention map, which allows integrating multiple stroke patterns into different spatial regions of the output image harmoniously. We demonstrate the effectiveness of our method, as well as generate comparable stylized images with multiple stroke patterns against the state-of-the-art methods.

Short videos are becoming key for media consumers exploring the TV and Internet. The production of short videos however remains costly. In this paper, we present a domain specific video summarization application with idiom adaptation that leverages multimedia content analysis and insights from cinematic and persuasive domains. From the back-end, content curators can push raw materials and the pre-processing algorithms will automatically extract the features and encode them as editing idioms. Users can create personalized video summaries based on these idioms. We have validated the effectiveness of the demonstration on a TVC data-set with over 600 videos, enabling production of domain specific video summaries with combinations of editing idioms. This approach has been put into trial in the testbed at Alibaba Wood.

As short videos become an increasingly popular form of storytelling, there is a growing demand for video summarization to convey information concisely with a subset of video frames. Some criteria such as interestingness and diversity are used by existing efforts to pick appropriate segments of content. However, there lacks a mechanism to infuse insights from cinematography and persuasion into this process. As a result, the results of the video summarization sometimes deviate from the original. In addition, the exploration of the vast design space to create customized video summaries is costly for video producer. To address these challenges, we propose a domain specific and idiom adaptive video summarization approach. Specifically, our approach first segments the input video and extracts high-level information from each segment. Such labels are used to represent a collection of idioms and summarization metrics as submodular components which users can combine to create personalized summary styles in a variety of ways. In order to identify the importance of the idioms and metrics in different domains, we leverage max margin learning. Experimental results have validated the effectiveness of our approach. We also plan to release a dataset containing over 600 videos with expert annotations which can benefit further research in this area.