Visual abductive reasoning (VAR) is a challenging task that requires AI systems to infer the most likely explanation for incomplete visual observations. While recent MLLMs develop strong general-purpose multimodal reasoning capabilities, they remain fall short in abductive inference, as compared to human beings. To bridge this gap, we draw inspiration from the interplay between verbal and pictorial abduction in human cognition, and propose to strengthen abduction of MLLMs by mimicking such dual-mode behavior. Concretely, we introduce AbductiveMLLM comprising of two synergistic components: REASONER and IMAGINER. The REASONER operates in the verbal domain. It first explores a broad space of possible explanations using a blind LLM and then prunes visually incongruent hypotheses based on cross-modal causal alignment. The remaining hypotheses are introduced into the MLLM as targeted priors, steering its reasoning toward causally coherent explanations. The IMAGINER, on the other hand, further guides MLLMs by emulating human-like pictorial thinking. It conditions a text-to-image diffusion model on both the input video and the REASONER’s output embeddings to “imagine” plausible visual scenes that correspond to verbal explanation, thereby enriching MLLMs' contextual grounding. The two components are trained jointly in an end-to-end manner. Experiments on standard VAR benchmarks show that AbductiveMLLM achieves state-of-the-art performance, consistently outperforming traditional solutions and advanced MLLMs.

Prompt tuning (PT) has emerged as a key to unlocking the power of visual-language models like CLIP for various downstream tasks. Predominant approaches learn a small set of task-relevant soft prompts by solving an image-class matching problem. Nevertheless, by optimizing merely with respect to class names, they face challenges in learning high performant prompts capable of capturing fine-grained, diverse characteristics of each class, and tends to overfit potentially biased distribution of base classes. In this work, we propose PTinCAS to tackle prompt tuning in a compact attribute space, driven by the premise that attributes offer detailed class interpretations and can facilitate transfer across related categories. Particularly, PTinCAS is grounded in two innovative designs. First, we create a compact attribute space by properly prompting large language models to generate factual descriptions about categories, which are subsequently clustered to form a concise attribute vocabulary. Second, we leverage attributes as a source of supervision in PT to transfer the inherent common sense knowledge in attributes to soft prompts. An object-aware visual prompting mechanism is developed to effortlessly highlight intended regions in the original image, which guides the model towards learning visual attributes associated with object regions rather than the background. We show that PTinCAS not only improves few-shot generalizability compared to existing PT methods, but also provides some level of inherent explainability that helps us understand why a class name is determined based on the attributes activated in an image.

The CLIP model has demonstrated significant advancements in aligning visual and language modalities through large-scale pre-training on image-text pairs, enabling strong zero-shot classification and retrieval capabilities on various domains. However, CLIP’s training remains computationally intensive, with high demands on both data processing and memory. To address these challenges, recent masking strategies have emerged, focusing on the selective removal of image patches to improve training efficiency. Although effective, these methods often compromise key semantic information, resulting in suboptimal alignment between visual features and text descriptions. In this work, we present a concise yet effective approach called Patch Generation-to-Selection (CLIP-PGS) to enhance CLIP’s training efficiency while preserving critical semantic con tent. Our method introduces a gradual masking process in which a small set of candidate patches is first pre-selected as potential mask regions. Then, we apply Sobel edge detection across the entire image to generate an edge mask that prioritizes the retention of the primary object areas. Finally, similarity scores between the candidate mask patches and their neighboring patches are computed, with optimal transport normalization refining the selection process to ensure a balanced similarity matrix. Our approach, CLIP-PGS, sets new state-of-the-art results in zero-shot classification and retrieval tasks, achieving superior performance in robustness evaluation and language compositionality benchmarks.

In modern urban environments, camera networks generate massive amounts of operational footage -- reaching petabytes each day -- making scalable video analytics essential for efficient processing. Many existing approaches adopt an SQL-based paradigm for querying such large-scale video databases; however, this constrains queries to rigid patterns with predefined semantic categories, significantly limiting analytical flexibility. In this work, we explore a language-driven video analytics paradigm aimed at enabling flexible and efficient querying of high-volume video data driven by natural language. Particularly, we build Lava, a system that accepts natural language queries and retrieves traffic targets across multiple levels of granularity and arbitrary categories. Lava comprises three main components: 1) a multi-armed bandit-based efficient sampling method for video segment-level localization; 2) a video-specific open-world detection module for object-level retrieval; and 3) a long-term object trajectory extraction scheme for temporal object association, yielding complete trajectories for object-of-interests. To support comprehensive evaluation, we further develop a novel benchmark by providing diverse, semantically rich natural language predicates and fine-grained annotations for multiple videos. Experiments on this benchmark demonstrate that Lava improves F1-scores for selection queries by 14% reduces MPAE for aggregation queries by 0.39, and achieves top-k precision of 86% while processing videos 9.6x faster than the most accurate baseline. Our code and dataset are available at https://github.com/yuyanrui/LAVA.

Reinforcement fine-tuning (RFT) has shown great promise in achieving humanlevel reasoning capabilities of Large Language Models (LLMs), and has recently been extended to MLLMs. Nevertheless, reasoning about videos, which is a fundamental aspect of human intelligence, remains a persistent challenge due to the complex logic, temporal and causal structures inherent in video data. To fill this gap, we propose VideoRFT, a novel approach that extends the RFT paradigm to cultivate human-like video reasoning capabilities in MLLMs. VideoRFT follows the standard two-stage scheme in RFT: supervised fine-tuning (SFT) with chain-of-thought (CoT) annotations, followed by reinforcement learning (RL) to improve generalization. A central challenge to achieve this in the video domain lies in the scarcity of large-scale, high-quality video CoT datasets. We address this by building a multi-expert-driven, cognition-inspired CoT curation pipeline. First, we devise a cognition-inspired prompting strategy to elicit a reasoning LLM to generate preliminary CoTs based solely on rich, structured, and literal representations of video content. Subsequently, these CoTs are revised by a MLLM conditioned on the actual video, ensuring visual consistency and reducing visual hallucinations. This pipeline results in two new datasets, i.e.VideoRFT-CoT-102K for SFT and VideoRFT-RL-310K for RL. To further strengthen the RL phase, we introduce a novel semantic-consistency reward that explicitly promotes the alignment between textual reasoning and visual evidence. This reward encourages the model to produce coherent, context-aware reasoning outputs grounded in visual input. Extensive experiments show that VideoRFT achieves state-of-the-art performance on six video reasoning benchmarks.

Prompt tuning (PT), as an emerging resource-efficient fine-tuning paradigm, has showcased remarkable effectiveness in improving the task-specific transferability of vision-language models. This paper delves into a previously overlooked information asymmetry issue in PT, where the visual modality mostly conveys more context than the object-oriented textual modality. Correspondingly, coarsely aligning these two modalities could result in the biased attention, driving the model to merely focus on the context area. To address this, we propose DAPT, an effective PT framework based on an intuitive decouple-before-align concept. First, we propose to explicitly decouple the visual modality into the foreground and background representation via exploiting coarse-and-fine visual segmenting cues, and then both of these decoupled patterns are aligned with the original foreground texts and the hand-crafted background classes, thereby symmetrically strengthening the modal alignment. To further enhance the visual concentration, we propose a visual pull-push regularization tailored for the foreground-background patterns, directing the original visual representation towards unbiased attention on the region-of-interest object. We demonstrate the power of architecture-free DAPT through few-shot learning, base-to-novel generalization, and data-efficient learning, all of which yield superior performance across prevailing benchmarks.

With the increasing popularity of autonomous driving systems and their applications in complex transportation scenarios, collaborative perception among multiple intelligent agents has become an important research direction. Existing single-agent multimodal fusion approaches are limited by their inability to leverage additional sensory data from nearby agents. In this article, we present the collaborative multimodal fusion network (CMMFNet) for distributed perception in multiagent systems. CMMFNet first extracts modality-specific features from LiDAR point clouds and camera images for each agent using dual-stream neural networks. To overcome the ambiguity in-depth prediction, we introduce a collaborative depth supervision module that projects dense fused point clouds onto image planes to generate more accurate depth ground truths. We then present modality-aware fusion strategies to aggregate homogeneous features across agents while preserving their distinctive properties. To align heterogeneous LiDAR and camera features, we introduce a modality consistency learning method. Finally, a transformer-based fusion module dynamically captures cross-modal correlations to produce a unified representation. Comprehensive evaluations on two extensive multiagent perception datasets, OPV2V and V2XSet, affirm the superiority of CMMFNet in detection performance, establishing a new benchmark in the field.

Remote sensing change detection is essential for environmental monitoring, urban planning, and related applications. However, current methods often struggle to capture long-range dependencies while maintaining computational efficiency. Although Transformers can effectively model global context, their quadratic complexity poses scalability challenges, and existing linear attention approaches frequently fail to capture intricate spatiotemporal relationships. Drawing inspiration from the recent success of Titans in language tasks, we present ChangeTitans, the Titans-based framework for remote sensing change detection. Specifically, we propose VTitans, the first Titans-based vision backbone that integrates neural memory with segmented local attention, thereby capturing long-range dependencies while mitigating computational overhead. Next, we present a hierarchical VTitans-Adapter to refine multi-scale features across different network layers. Finally, we introduce TS-CBAM, a two-stream fusion module leveraging cross-temporal attention to suppress pseudo-changes and enhance detection accuracy. Experimental evaluations on four benchmark datasets (LEVIR-CD, WHU-CD, LEVIR-CD+, and SYSU-CD) demonstrate that ChangeTitans achieves state-of-the-art results, attaining 84.36% IoU and 91.52% F1-score on LEVIR-CD, while remaining computationally competitive. Our code and model are available at https://github.com/ChangeTitans/ChangeTitans.

The substantial successes achieved by diffusion probabilistic models have prompted the study of their employment in resource-limited scenarios. Pruning methods have been proven effective in compressing discriminative models relying on the correlation between training losses and model performances. However, diffusion models employ an iterative process for generating high-quality images, leading to a breakdown of such connections. To address this challenge, we propose a simple yet effective method, named NiCI-Pruning (Noise in Clean Image Pruning), for the compression of diffusion models. NiCI-Pruning capitalizes the noise predicted by the model based on clean image inputs, favoring it as a feature for establishing reconstruction losses. Accordingly, Taylor expansion is employed for the proposed reconstruction loss to evaluate the parameter importance effectively. Moreover, we propose an interval sampling strategy that incorporates a timestep-weighted schema, alleviating the risk of misleading information obtained at later timesteps. We provide comprehensive experimental results to affirm the superiority of our proposed approach. Notably, our method achieves a remarkable average reduction of 30.4% in FID score increase across five different datasets compared to the state-of-the-art diffusion pruning method at equivalent pruning rates. Our code and models have been made available at https://github.com/NUST-Machine-Intelligence-Laboratory/NiCI-Pruning.

The goal of this paper is to alleviate the training cost for few-shot semantic segmentation (FSS) models. Despite that FSS in nature improves model generalization to new concepts using only a handful of test exemplars, it relies on strong supervision from a considerable amount of labeled training data for base classes. However, collecting pixel-level annotations is notoriously expensive and time-consuming, and small-scale training datasets convey low information density that limits test-time generalization. To resolve the issue, we take a pioneering step towards label-efficient training of FSS models from fully unlabeled training data, or additionally a few labeled samples to enhance the performance. This motivates an approach based on a novel unsupervised meta-training paradigm. In particular, the approach first distills pre-trained unsupervised pixel embedding into compact semantic clusters from which a massive number of pseudo meta-tasks is constructed. To mitigate the noise in the pseudo meta-tasks, we further advocate a robust Transformer-based FSS model with a novel prototype-based cross-attention design. Extensive experiments have been conducted on two standard benchmarks, i.e., PASCAL-5i and COCO-20i, and the results show that our method produces impressive performance without any annotations, and is comparable to fully supervised competitors even using only 20% of the annotations. Our code is available at: https://github.com/SSSKYue/UMTFSS.

This paper addresses the problem of on-road object importance estimation, which utilizes video sequences captured from the driver's perspective as the input. Although this problem is significant for safer and smarter driving systems, the exploration of this problem remains limited. On one hand, publicly-available large-scale datasets are scarce in the community. To address this dilemma, this paper contributes a new large-scale dataset named Traffic Object Importance (TOI). On the other hand, existing methods often only consider either bottom-up feature or single-fold guidance, leading to limitations in handling highly dynamic and diverse traffic scenarios. Different from existing methods, this paper proposes a model that integrates multi-fold top-down guidance with the bottom-up feature. Specifically, three kinds of top-down guidance factors (i.e., driver intention, semantic context, and traffic rule) are integrated into our model. These factors are important for object importance estimation, but none of the existing methods simultaneously consider them. To our knowledge, this paper proposes the first on-road object importance estimation model that fuses multi-fold top-down guidance factors with bottom-up feature. Extensive experiments demonstrate that our model outperforms state-of-the-art methods by large margins, achieving 23.1% Average Precision (AP) improvement compared with the recently proposed model (i.e., Goal).

Humans are able to recognize structured relations in observation, allowing us to decompose complex scenes into simpler parts and abstract the visual world at multiple levels. However, such hierarchical reasoning ability of human perception remains largely unexplored in current literature of semantic segmentation. Existing works are often aware of flatten labels and distinguish all the semantic categories exclusively for each pixel. In this work, we instead address hierarchical semantic segmentation (HSS), with the aim of providing a structured, pixel-wise description of visual observation in terms of a class hierarchy. We deviseHssn, a general HSS framework that tackles two critical issues in this task:i)how to efficiently adapt existing hierarchy-agnostic segmentation networks to the HSS setting, andii)how to leverage the class hierarchy to regularize HSS network learning. To addressi),Hssndirectly casts HSS as a pixel-wise multi-label classification task, only bringing minimal architecture change to current segmentation models. To solveii),Hssnfirst explores inherent properties of the hierarchy as a training objective, which enforces segmentation predictions to obey the hierarchy structure. Furthermore, with a set of hierarchy-induced margin constraints,Hssnefficiently reshapes the learned pixel embedding space, so as to generate hierarchy-aware pixel representations and facilitate structured segmentation eventually. Building uponHssn, we further exploit the mutual exclusion relation between semantic labels and strengthen the margin based regularization strategy with more meaningful constrains, leading toHssn+, a more effective framework for HSS. We conduct extensive experiments on six semantic segmentation datasets (i.e., Mapillary Vistas 2.0, Cityscapes, LIP, PASCAL-Person-Part, PASCAL-Part-58, and PASCAL-Part-108), with different class hierarchies, network architectures, and backbones, and the results confirm the generalization and superiority of our algorithms.

This work studies the problem of image semantic segmentation. Current approaches focus mainly on mining "local" context, i.e., dependencies between pixels within individual images, by specifically-designed, context aggregation modules (e.g., dilated convolution, neural attention) or structure-aware optimization objectives (e.g., IoU-like loss). However, they ignore "global" context of the training data, i.e., rich semantic relations between pixels across different images. Inspired by recent advance in unsupervised contrastive representation learning, we propose a pixel-wise contrastive algorithm, dubbed as PiCo, for semantic segmentation in the fully supervised learning setting. The core idea is to enforce pixel embeddings belonging to a same semantic class to be more similar than embeddings from different classes. It raises a pixel-wise metric learning paradigm for semantic segmentation, by explicitly exploring the structures of labeled pixels, which were rarely studied before. Our training algorithm is compatible with modern segmentation solutions without extra overhead during testing. We experimentally show that, with famous segmentation models (i.e., DeepLabV3, HRNet, OCRNet, SegFormer, Segmenter, MaskFormer) and backbones (i.e., MobileNet, ResNet, HRNet, MiT, ViT), our algorithm brings consistent performance improvements across diverse datasets (i.e., Cityscapes, ADE20 K, PASCAL-Context, COCO-Stuff, CamVid). We expect that this work will encourage our community to rethink the current de facto training paradigm in semantic segmentation.

Deep learning based semantic segmentation solutions have yielded compelling results over the preceding decade. They encompass diverse network architectures (FCN based or attention based), along with various mask decoding schemes (parametric softmax based or pixel-query based). Despite the divergence, they can be grouped within a unified framework by interpreting the softmax weights or query vectors as learnable class prototypes. In light of this prototype view, we reveal inherent limitations within the parametric segmentation regime, and accordingly develop a nonparametric alternative based on non-learnable prototypes. In contrast to previous approaches that entail the learning of a single weight/query vector per class in a fully parametric manner, our approach represents each class as a set of non-learnable prototypes, relying solely upon the mean features of training pixels within that class. The pixel-wise prediction is thus achieved by nonparametric nearest prototype retrieving. This allows our model to directly shape the pixel embedding space by optimizing the arrangement between embedded pixels and anchored prototypes. It is able to accommodate an arbitrary number of classes with a constant number of learnable parameters. Through empirical evaluation with FCN based and Transformer based segmentation models (i.e., HRNet, Swin, SegFormer, Mask2Former) and backbones (i.e., ResNet, HRNet, Swin, MiT), our nonparametric framework shows superior performance on standard segmentation datasets (i.e., ADE20 K, Cityscapes, COCO-Stuff), as well as in large-vocabulary semantic segmentation scenarios. We expect that this study will provoke a rethink of the current de facto semantic segmentation model design.

Federated learning is a distributed paradigm that allows multiple parties to collaboratively train deep learning models without direct exchange of raw data. Nevertheless, the inherent non-independent and identically distributed (non-i.i.d.) nature of data distribution among clients results in significant degradation of the acquired model. The primary goal of this study is to develop a robust federated learning algorithm to addressfeature shiftin clients’ samples, potentially arising from a range of factors such as acquisition discrepancies in medical imaging. To reach this goal, we first propose federated feature augmentation (FedFA$^{l}$), a novel feature augmentation technique tailored for federated learning.FedFA$^{l}$is based on a crucial insight that each client's data distribution can be characterized by first-/second-order statistics (a.k.a., mean and standard deviation) of latent features; and it is feasible to manipulate these local statisticsglobally, i.e., based on information in the entire federation, to let clients have a better sense of the global distribution across clients. Grounded on this insight, we propose to augment each local feature statistic based on a normal distribution, wherein the mean corresponds to the original statistic, and the variance defines the augmentation scope. Central toFedFA$^{l}$is the determination of a meaningful Gaussian variance, which is accomplished by taking into account not only biased data of each individual client, but also underlying feature statistics represented by all participating clients. Beyond consideration oflow-orderstatistics inFedFA$^{l}$, we propose a federated feature alignment component (FedFA$^{h}$) that exploitshigher-orderfeature statistics to gain a more detailed understanding of local feature distribution and enables explicit alignment of augmented features in different clients to promote more consistent feature learning. CombiningFedFA$^{l}$andFedFA$^{h}$yields our full approachFedFA$+$+.FedFA$+$+is non-parametric, incurs negligible additional communication costs, and can be seamlessly incorporated into popular CNN and Transformer architectures. We offer rigorous theoretical analysis, as well as extensive empirical justifications to demonstrate the effectiveness of the algorithm.

In this paper, we propose nsDB, a novel neuro-symbolic database system that integrates neural and symbolic system architectures natively to address the weaknesses of each, providing a strong database capable of data managing, model learning, and complex analytical query processing over multi-modal data. We employ a real-world NBA data analytical query as an example to illustrate the functionality of each component in nsDB and highlight the research challenges to build it. We then present the key design principles and our preliminary attempts to address them. In a nutshell, we envision that the next generation database system nsDB integrates the complex neural system with the simple symbolic system. Undoubtedly, nsDB will serve as a bridge between databases with AI models, which abstracts away the AI complexities but allows end users to enjoy the strong capabilities of them. We are in the early stages of the journey to build nsDB, there are many opening challenges, e.g., in-database model training, multi-objective query optimization, and database agent development. We hope the researchers from different communities (e.g., system, architecture, database, artificial intelligence) could tackle them together.

Currently, the success of image processing relies heavily on large well-annotated datasets. However, collecting and labeling video data are significantly more labor-intensive, posing major challenges for training video algorithms and limiting their practical applications. While label-efficient techniques for image data have advanced, solutions for video data are still emerging. Unlabeled video data, with their inherent structured nature, offer valuable assets for label-efficient learning. Unlike image data, video data naturally captures realistic transformations, providing rich samples for learning. Moreover, from a border perspective, video tasks hold great potential for applications like autonomous driving and video surveillance but present unique challenges due to the need to understand both spatial and temporal aspects. Leveraging label-efficient learning is essential for comprehensively understanding visual content and enabling a wide range of real-world video applications. This Special Issue on “Label-Efficient Learning for Video Data” seeks to advance research in this area, offering new insights and solutions to benefit both researchers and practitioners.

Point cloud semantic segmentation is a fundamental task in 3D scene understanding and has recently achieved remarkable progress. The success of existing approaches is attributed to recent advanced deep networks for point clouds and the availability of a large amount of labeled training data. However, creating such fully annotated training datasets for supervised point cloud semantic segmentation methods is a time-consuming and labor-intensive process, which increases the difficulty of extending supervised approaches to new application scenarios. To alleviate the data-hungry nature of deep learning, we propose PCL, the point contrast and labeling framework for weakly supervised point cloud semantic segmentation with small percentages of point-level annotations. The core idea of this method is to exploit contrastive learning to help learn a larger number of discriminative feature representations with limited annotations. By introducing two types of contrastive relationships, cross-sample point contrast and low-level similarity-based point contrast, our proposed framework can directly regularize the learned feature space, considering not only the low-level similarity within each point cloud but also the discriminative semantics within and across point clouds on both labeled and unlabeled points via pseudo labels. In addition, we propose a pseudo label refinery module to generate robust and reliable pseudo labels online, reducing the negative impact of incorrect pseudo labels. Our method achieves state-of-the-art performance on a diverse set of label-efficient semantic segmentation tasks.

The objective of this paper is self-supervised learning of video object segmentation. We develop a unified framework which simultaneously models cross-frame dense correspondence for locally discriminative feature learning and embeds object-level context for target-mask decoding. As a result, it is able to directly learn to perform mask-guided sequential segmentation from unlabeled videos, in contrast to previous efforts usually relying on an oblique solution - cheaply “copying” labels according to pixel-wise correlations. Concretely, our algorithm alternates between i) clustering video pixels for creating pseudo segmentation labels ex nihilo; and ii) utilizing the pseudo labels to learn mask encoding and decoding for VOS. Unsupervised correspondence learning is further incorporated into this self-taught, mask embedding scheme, so as to ensure the generic nature of the learnt representation and avoid cluster degeneracy. Our algorithm sets state-of-the-arts on two standard benchmarks (i.e., DAVIS17and YouTube-VOS), narrowing the gap between self- and fully-supervised VOS, in terms of both performance and network architecture design.

This paper proposes an anchor-based deformation model, namely AnchorDEF, to predict 3D garment animation from a body motion sequence. It deforms a garment mesh template by a mixture of rigid transformations with extra nonlinear displacements. A set of anchors around the mesh surface is introduced to guide the learning of rigid transformation matrices. Once the anchor transformations are found, per-vertex nonlinear displacements of the garment template can be regressed in a canonical space, which reduces the complexity of deformation space learning. By explicitly constraining the transformed anchors to satisfy the consistencies of position, normal and direction, the physical meaning of learned anchor transformations in space is guaranteed for better generalization. Furthermore, an adaptive anchor updating is proposed to optimize the anchor position by being aware of local mesh topology for learning representative anchor transformations. Qualitative and quantitative experiments on different types of garments demonstrate that AnchorDEF achieves the state-of-the-art performance on 3D garment deformation prediction in motion, especially for loose-fitting garments.

This paper presents Core, a conceptually simple, effective and communication-efficient model for multi-agent cooperative perception. It addresses the task from a novel perspective of cooperative reconstruction, based on two key insights: 1) cooperating agents together provide a more holistic observation of the environment, and 2) the holistic observation can serve as valuable supervision to explicitly guide the model learning how to reconstruct the ideal observation based on collaboration. Core instantiates the idea with three major components: a compressor for each agent to create more compact feature representation for efficient broadcasting, a lightweight attentive collaboration component for cross-agent message aggregation, and a reconstruction module to reconstruct the observation based on aggregated feature representations. This learning-to-reconstruct idea is task-agnostic, and offers clear and reasonable supervision to inspire more effective collaboration, eventually promoting perception tasks. We validate Core on two large-scale multi-agent percetion dataset, OPV2V and V2X-Sim, in two tasks, i.e., 3D object detection and semantic segmentation. Results demonstrate that Core achieves state-of-the-art performance, and is more communication-efficient.

We present CLUSTSEG, a general, transformer-based framework that tackles different image segmentation tasks ($i.e.,$ superpixel, semantic, instance, and panoptic) through a unified, neural clustering scheme. Regarding queries as cluster centers, CLUSTSEG is innovative in two aspects: 1) cluster centers are initialized in heterogeneous ways so as to pointedly address task-specific demands ($e.g.,$ instance- or category-level distinctiveness), yet without modifying the architecture; and 2) pixel-cluster assignment, formalized in a cross-attention fashion, is alternated with cluster center update, yet without learning additional parameters. These innovations closely link CLUSTSEG to EM clustering and make it a transparent and powerful framework that yields superior results across the above segmentation tasks.

This paper studies the problem of weakly open-vocabulary semantic segmentation (WOVSS), which learns to segment objects of arbitrary classes using mere image-text pairs. Existing works turn to enhance the vanilla vision transformer by introducing explicit grouping recognition, i.e., employing several group tokens/centroids to cluster the image tokens and perform the group-text alignment. Nevertheless, these methods suffer from a granularity inconsistency regarding the usage of group tokens, which are aligned in the all-to-one v.s. one-to-one manners during the training and inference phases, respectively. We argue that this discrepancy arises from the lack of elaborate supervision for each group token. To bridge this granularity gap, this paper explores explicit supervision for the group tokens from the prototypical knowledge. To this end, this paper proposes the non-learnable prototypical regularization (NPR) where non-learnable prototypes are estimated from source features to serve as supervision and enable contrastive matching of the group tokens. This regularization encourages the group tokens to segment objects with less redundancy and capture more comprehensive semantic regions, leading to increased compactness and richness. Based on NPR, we propose the prototypical guidance segmentation network (PGSeg) that incorporates multi-modal regularization by leveraging prototypical sources from both images and texts at different levels, progressively enhancing the segmentation capability with diverse prototypical patterns. Experimental results show that our proposed method achieves state-of-the-art performance on several benchmark datasets.

Despite recent progress of automatic medical image segmentation techniques, fully automatic results usually fail to meet clinically acceptable accuracy, thus typically require further refinement. To this end, we propose a novel Volumetric Memory Network, dubbed as VMN, to enable segmentation of 3D medical images in an interactive manner. Provided by user hints on an arbitrary slice, a 2D interaction network is firstly employed to produce an initial 2D segmentation for the chosen slice. Then, the VMN propagates the initial segmentation mask bidirectionally to all slices of the entire volume. Subsequent refinement based on additional user guidance on other slices can be incorporated in the same manner. To facilitate smooth human-in-the-loop segmentation, a quality assessment module is introduced to suggest the next slice for interaction based on the segmentation quality of each slice produced in the previous round. Our VMN demonstrates two distinctive features: First, the memory-augmented network design offers our model the ability to quickly encode past segmentation information, which will be retrieved later for the segmentation of other slices; Second, the quality assessment module enables the model to directly estimate the quality of each segmentation prediction, which allows for an active learning paradigm where users preferentially label the lowest-quality slice for multi-round refinement. The proposed network leads to a robust interactive segmentation engine, which can generalize well to various types of user annotations (e.g., scribble, bounding box, extreme clicking). Extensive experiments have been conducted on three public medical image segmentation datasets (i.e., MSD, KiTS19, CVC-ClinicDB), and the results clearly confirm the superiority of our approach in comparison with state-of-the-art segmentation models. The code is made publicly available at https://github.com/0liliulei/Mem3D.

We explore the task of language-guided video segmentation (LVS). Previous algorithms mostly adopt 3D CNNs to learn video representation, struggling to capture long-term context and easily suffering from visual-linguistic misalignment. In light of this, we present Locater (local-global context aware Transformer), which augments the Transformer architecture with a finite memory so as to query the entire video with the language expression in an efficient manner. The memory is designed to involve two components - one for persistently preserving global video content, and one for dynamically gathering local temporal context and segmentation history. Based on the memorized local-global context and the particular content of each frame, Locater holistically and flexibly comprehends the expression as an adaptive query vector for each frame. The vector is used to query the corresponding frame for mask generation. The memory also allows Locater to process videos with linear time complexity and constant size memory, while Transformer-style self-attention computation scales quadratically with sequence length. To thoroughly examine the visual grounding capability of LVS models, we contribute a new LVS dataset, A2D-S +, which is built upon A2D-S dataset but poses increased challenges in disambiguating among similar objects. Experiments on three LVS datasets and our A2D-S + show that Locater outperforms previous state-of-the-arts. Further, we won the 1st place in the Referring Video Object Segmentation Track of the 3rd Large-scale Video Object Segmentation Challenge, where Locater served as the foundation for the winning solution.

Video segmentation-partitioning video frames into multiple segments or objects-plays a critical role in a broad range of practical applications, from enhancing visual effects in movie, to understanding scenes in autonomous driving, to creating virtual background in video conferencing. Recently, with the renaissance of connectionism in computer vision, there has been an influx of deep learning based approaches for video segmentation that have delivered compelling performance. In this survey, we comprehensively review two basic lines of research - generic object segmentation (of unknown categories) in videos, and video semantic segmentation - by introducing their respective task settings, background concepts, perceived need, development history, and main challenges. We also offer a detailed overview of representative literature on both methods and datasets. We further benchmark the reviewed methods on several well-known datasets. Finally, we point out open issues in this field, and suggest opportunities for further research. We also provide a public website to continuously track developments in this fast advancing field: https://github.com/tfzhou/VS-Survey.

In this work, we study the challenging problem of instance-aware human body part parsing. We introduce a new bottom-up regime which achieves the task through learning category-level human semantic segmentation as well as multi-person pose estimation in a joint and end-to-end manner. The output is a compact, efficient and powerful framework that exploits structural information over different human granularities and eases the difficulty of person partitioning. Specifically, a dense-to-sparse projection field, which allows explicitly associating dense human semantics with sparse keypoints, is learnt and progressively improved over the network feature pyramid for robustness. Then, the difficult pixel grouping problem is cast as an easier, multi-person joint assembling task. By formulating joint association as maximum-weight bipartite matching, we develop two novel algorithms based on projected gradient descent and unbalanced optimal transport, respectively, to solve the matching problem differentiablly. These algorithms make our method end-to-end trainable and allow back-propagating the grouping error to directly supervise multi-granularity human representation learning. This is significantly distinguished from current bottom-up human parsers or pose estimators which require sophisticated post-processing or heuristic greedy algorithms. Extensive experiments on three instance-aware human parsing datasets (i.e., MHP-v2, DensePose-COCO, PASCAL-Person-Part) demonstrate that our approach outperforms most existing human parsers with much more efficient inference. Our code is available at https://github.com/tfzhou/MG-HumanParsing.

Learning pyramidal feature representations is important for many dense prediction tasks (e.g., object detection, semantic segmentation) that demand multi-scale visual understanding. Feature Pyramid Network (FPN) is a well-known architecture for multi-scale feature learning, however, intrinsic weaknesses in feature extraction and fusion impede the production of informative features. This work addresses the weaknesses of FPN through a novel tripartite feature enhanced pyramid network (TFPN), with three distinct and effective designs. First, we develop a feature reference module with lateral connections to adaptively extract bottom-up features with richer details for feature pyramid construction. Second, we design a feature calibration module between adjacent layers that calibrates the upsampled features to be spatially aligned, allowing for feature fusion with accurate correspondences. Third, we introduce a feature feedback module in FPN, which creates a communication channel from the feature pyramid back to the bottom-up backbone and doubles the encoding capacity, enabling the entire architecture to generate incrementally more powerful representations. The TFPN is extensively evaluated over four popular dense prediction tasks, i.e., object detection, instance segmentation, panoptic segmentation, and semantic segmentation. The results demonstrate that TFPN consistently and significantly outperforms the vanilla FPN. Our code is available at https://github.com/jamesliang819.

Recently, numerous algorithms have been developed to tackle the problem of light field super-resolution (LFSR), i.e., super-resolving low-resolution light fields to gain high-resolution views. Despite delivering encouraging results, these approaches are all convolution-based, and are naturally weak in global relation modeling of sub-aperture images necessarily to characterize the inherent structure of light fields. In this paper, we put forth a novel formulation built upon Transformers, by treating LFSR as a sequence-to-sequence reconstruction task. In particular, our model regards sub-aperture images of each vertical or horizontal angular view as a sequence, and establishes long-range geometric dependencies within each sequence via a spatial-angular locally-enhanced self-attention layer, which maintains the locality of each sub-aperture image as well. Additionally, to better recover image details, we propose a detail-preserving Transformer (termed as DPT), by leveraging gradient maps of light field to guide the sequence learning. DPT consists of two branches, with each associated with a Transformer for learning from an original or gradient image sequence. The two branches are finally fused to obtain comprehensive feature representations for reconstruction. Evaluations are conducted on a number of light field datasets, including real-world scenes and synthetic data. The proposed method achieves superior performance comparing with other state-of-the-art schemes. Our code is publicly available at: https://github.com/BITszwang/DPT.

Humans are able to recognize structured relations in observation, allowing us to decompose complex scenes into simpler parts and abstract the visual world in multiple levels. However, such hierarchical reasoning ability of human perception remains largely unexplored in current literature of semantic segmentation. Existing work is often aware of flatten labels and predicts target classes exclusively for each pixel. In this paper, we instead address hierarchical semantic segmentation (HSS), which aims at structured, pixel-wise description of visual observation in terms of a class hierarchy. We devise HSSN, a general HSS framework that tackles two critical issues in this task: i) how to efficiently adapt existing hierarchy-agnostic segmentation networks to the HSS setting, and ii) how to leverage the hierarchy information to regularize HSS network learning. To address i), HSSN directly casts HSS as a pixel-wise multi-label classification task, only bringing minimal architecture change to current segmentation models. To solve ii), HSSN first explores inherent properties of the hierarchy as a training objective, which enforces segmentation predictions to obey the hierarchy structure. Further, with hierarchy-induced margin constraints, HSSNreshapes the pixel embedding space, so as to generate well-structured pixel representations and improve segmentation eventually. We conduct experiments on four semantic segmentation datasets (i.e., Mapillary Vistas 2.0, City-scapes, LIP, and PASCAL-Person-Part), with different class hierarchies, segmentation network architectures and backbones, showing the generalization and superiority of HSSN.

Our target is to learn visual correspondence from unlabeled videos. We develop Liir, a locality-aware inter-and intra-video reconstruction method that fills in three missing pieces, i.e., instance discrimination, location awareness, and spatial compactness, of self-supervised correspondence learning puzzle. First, instead of most existing efforts focusing on intra-video self-supervision only, we exploit cross-video affinities as extra negative samples within a unified, inter-and intra-video reconstruction scheme. This enables instance discriminative representation learning by contrasting desired intra-video pixel association against negative inter-video correspondence. Second, we merge position information into correspondence matching, and design a position shifting strategy to remove the side-effect of position encoding during inter-video affinity computation, making our Liir location-sensitive. Third, to make full use of the spatial continuity nature of video data, we impose a compactness-based constraint on correspondence matching, yielding more sparse and reliable solutions. The learned representation surpasses self-supervised state-of-the-arts on label propagation tasks including objects, semantic parts, and keypoints.

Abductive reasoning seeks the likeliest possible explanation for partial observations. Although abduction is frequently employed in human daily reasoning, it is rarely explored in computer vision literature. In this paper, we propose a new task and dataset, Visual Abductive Reasoning (VAR), for examining abductive reasoning ability of machine intelligence in everyday visual situations. Given an incomplete set of visual events, AI systems are required to not only describe what is observed, but also infer the hypothesis that can best explain the visual premise. Based on our large-scale VAR dataset, we devise a strong baseline model, REASONER (causal-and-cascaded reasoning Transformer). First, to capture the causal structure of the observations, a contextualized directional position embedding strategy is adopted in the encoder, that yields discriminative represen-tations for the premise and hypothesis. Then, multiple de-coders are cascaded to generate and progressively refine the premise and hypothesis sentences. The prediction scores of the sentences are used to guide cross-sentence information flow in the cascaded reasoning procedure. Our VAR bench-marking results show that REASONER surpasses many famous video-language models, while still being far behind human performance. This work is expected to foster future efforts in the reasoning-beyond-observation paradigm.

Prevalent semantic segmentation solutions, despite their different network designs (FCN based or attention based) and mask decoding strategies (parametric softmax based or pixel-query based), can be placed in one category, by considering the softmax weights or query vectors as learnable class prototypes. In light of this prototype view, this study uncovers several limitations of such parametric segmentation regime, and proposes a nonparametric alternative based on non-learnable prototypes. Instead of prior methods learning a single weight/query vector for each class in a fully parametric manner, our model represents each class as a set of non-learnable prototypes, relying solely on the mean fea-tures of several training pixels within that class. The dense prediction is thus achieved by nonparametric nearest prototype retrieving. This allows our model to directly shape the pixel embedding space, by optimizing the arrangement between embedded pixels and anchored prototypes. It is able to handle arbitrary number of classes with a constant amount of learnable parameters. We empirically show that, with FCN based and attention based segmentation models (i.e., HR-Net, Swin, SegFormer) and backbones (i.e., ResNet, HRNet, Swin, MiT), our nonparametric framework yields compel-ling results over several datasets (i.e., ADE20K, Cityscapes, COCO-Stuff), and performs well in the large-vocabulary situation. We expect this work will provoke a rethink of the current de facto semantic segmentation model design.

Learning semantic segmentation from weakly-labeled (e.g., image tags only) data is challenging since it is hard to infer dense object regions from sparse semantic tags. Despite being broadly studied, most current efforts directly learn from limited semantic annotations carried by individual image or image pairs, and struggle to obtain integral localization maps. Our work alleviates this from a novel perspective, by exploring rich semantic contexts synergistically among abundant weakly-labeled training data for network learning and inference. In particular, we propose regional semantic contrast and aggregation (RCA). RCA is equipped with a regional memory bank to store massive, diverse object patterns appearing in training data, which acts as strong support for exploration of dataset-level semantic structure. Particularly, we propose i) semantic contrast to drive network learning by contrasting massive categorical object regions, leading to a more holistic object pattern understanding, and ii) semantic aggregation to gather diverse relational contexts in the memory to enrich semantic repre-sentations. In this manner, RCA earns a strong capability of fine-grained semantic understanding, and eventually establishes new state-of-the-art results on two popular benchmarks, i.e., PASCAL VOC 2012 and COCO 2014.

This paper solves the problem of learning image semantic segmentation using image-level supervision. The task is promising in terms of reducing annotation efforts, yet extremely challenging due to the difficulty to directly associate high-level concepts with low-level appearance. While current efforts handle each concept independently, we take a broader perspective to harvest implicit, holistic structures of semantic concepts, which express valuable prior knowledge for accurate concept grounding. This raises multi-granular semantic mining, a new formalism allowing flexible specification of complex relations in the label space. In particular, we propose a heterogeneous graph neural network (Hgnn) to model the heterogeneity of multi-granular semantics within a set of input images. The Hgnn consists of two types of sub-graphs: 1) an external graph characterizes the relations across different images to mine inter-image contexts; and for each image, 2) an internal graph is constructed to mine inter-class semantic dependencies within each individual image. Through heterogeneous graph learning, our Hgnn is able to land a comprehensive understanding of object patterns, leading to more accurate semantic concept grounding. Extensive experimental results show that Hgnn outperforms the current state-of-the-art approaches on the popular PASCAL VOC 2012 and COCO 2014 benchmarks. Our code is available at: https://github.com/maeve07/HGNN.git.

It is quite laborious and costly to manually label LiDAR point cloud data for training high-quality 3D object detectors. This work proposes a weakly supervised framework which allows learning 3D detection from a few weakly annotated examples. This is achieved by a two-stage architecture design. Stage-1 learns to generate cylindrical object proposals under inaccurate and inexact supervision, obtained by our proposed BEV center-click annotation strategy, where only the horizontal object centers are click-annotated in bird's view scenes. Stage-2 learns to predict cuboids and confidence scores in a coarse-to-fine, cascade manner, under incomplete supervision, i.e., only a small portion of object cuboids are precisely annotated. With KITTI dataset, using only 500 weakly annotated scenes and 534 precisely labeled vehicle instances, our method achieves 86-97 percent the performance of current top-leading, fully supervised detectors (which require 3,712 exhaustively annotated scenes with 15,654 instances). More importantly, with our elaborately designed network architecture, our trained model can be applied as a 3D object annotator, supporting both automatic and active (human-in-the-loop) working modes. The annotations generated by our model can be used to train 3D object detectors, achieving over 95 percent of their original performance (with manually labeled training data). Our experiments also show our model's potential in boosting performance when given more training data. The above designs make our approach highly practical and open-up opportunities for learning 3D detection at reduced annotation cost.

Modeling the human structure is central for human parsing that extracts pixel-wise semantic information from images. We start with analyzing three types of inference processes over the hierarchical structure of human bodies: direct inference (directly predicting human semantic parts using image information), bottom-up inference (assembling knowledge from constituent parts), and top-down inference (leveraging context from parent nodes). We then formulate the problem as a compositional neural information fusion (CNIF) framework, which assembles the information from the three inference processes in a conditional manner, i.e., considering the confidence of the sources. Based on CNIF, we further present a part-relation-aware human parser (PRHP), which precisely describes three kinds of human part relations, i.e., decomposition, composition, and dependency, by three distinct relation networks. Expressive relation information can be captured by imposing the parameters in the relation networks to satisfy specific geometric characteristics of different relations. By assimilating generic message-passing networks with their edge-typed, convolutional counterparts, PRHP performs iterative reasoning over the human body hierarchy. With these efforts, PRHP provides a more general and powerful form of CNIF, and lays the foundation for more sophisticated and flexible human relation patterns of reasoning. Experiments on five datasets demonstrate that our two human parsers outperform the state-of-the-arts in all cases.

This paper addresses the task of detecting and recognizing human-object interactions (HOI) in images. Considering the intrinsic complexity and structural nature of the task, we introduce a cascaded parsing network (CP-HOI) for a multi-stage, structured HOI understanding. At each cascade stage, an instance detection module progressively refines HOI proposals and feeds them into a structured interaction reasoning module. Each of the two modules is also connected to its predecessor in the previous stage, enabling efficient cross-stage information propagation. The structured interaction reasoning module is built upon a graph parsing neural network (GPNN), which efficiently models potential HOI structures as graphs and mines rich context for comprehensive relation understanding. In particular, GPNN infers a parse graph that i) interprets meaningful HOI structures by a learnable adjacency matrix, and ii) predicts action (edge) labels. Within an end-to-end, message-passing framework, GPNN blends learning and inference, iteratively parsing HOI structures and reasoning HOI representations (i.e., instance and relation features). Further beyond relation detection at a bounding-box level, we make our framework flexible to perform fine-grained pixel-wise relation segmentation; this provides a new glimpse into better relation modeling. A preliminary version of our CP-HOI model reached 1stplace in the ICCV2019 Person in Context Challenge, on both relation detection and segmentation. In addition, our CP-HOI shows promising results on two popular HOI recognition benchmarks,i.e., V-COCO and HICO-DET.

Current super-resolution networks typically reduce network parameters and multiadds operations by designing lightweight structures, but lightening the convolution layer is often ignored. In this work, we observe that$3 \times 3$convolutions occupy a high percentage of network parameters in most lightweight super-resolution networks. This motivates us to consider lightening super-resolution networks by replacing$3 \times 3$convolutions with lightweight convolutions, while maintaining the performance. To achieve this, we propose a lightweight convolution layer named contextual transformation layer (CTL). It can yield efficient contextual features through a context feature extraction module and enrich extracted contextual features through a context feature transformation module. Based on CTLs, we build a lightweight super-resolution network called contextual transformation network (CTN) for remote-sensing image super-resolution. Specifically, we use two CTLs to construct a contextual transformation block (CTB) for hierarchical feature learning. Interleaved with a CTB, a context enhancement module (CEM) is employed to enhance the extracted feature representations. All extracted features are processed by a contextual feature aggregation module for final remote-sensing image super-resolution. Extensive experiments are performed on a remote-sensing image super-resolution benchmark named UC Merced. Our method achieves superior results to the other state-of-the-art methods. To demonstrate the generalization ability of our CTL, we extend our CTN to two relevant tasks: natural image super-resolution and natural image denoising. Experimental results on natural image super-resolution benchmarks (i.e., Set5, Set14, B100, Urban100, and Manga109) and natural image denoising benchmarks (i.e., SIDD and DND) further prove the superiority of our method. Our code is publicly available athttps://github.com/BITszwang/CTNet.

The success of current deep saliency models heavily depends on large amounts of annotated human fixation data to fit the highly non-linear mapping between the stimuli and visual saliency. Such fully supervised data-driven approaches are annotation-intensive and often fail to consider the underlying mechanisms of visual attention. In contrast, in this paper, we introduce a model based on various cognitive theories of visual saliency, which learns visual attention patterns in a weakly supervised manner. Our approach incorporates insights from cognitive science as differentiable submodules, resulting in a unified, end-to-end trainable framework. Specifically, our model encapsulates the following important components motivated from biological vision. (a) As scene semantics are closely related to visually attentive regions, our model encodes discriminative spatial information for scene understanding through spatial visual semantics embedding. (b) To model the objectness factors in visual attention deployment, we incorporate object-level semantics embedding and object relation information. (c) Considering the "winner-take-all" mechanism in visual stimuli processing, we model the competition mechanism among objects with softmax based neural attention. (d) Lastly, a conditional center prior is learned to mimic the spatial distribution bias of visual attention. Furthermore, we propose novel loss functions to utilize supervision cues from image-level semantics, saliency prior knowledge, and self-information compression. Experiments show that our method achieves promising results, and even outperforms many of its fully supervised counterparts. Overall, our weakly supervised saliency method makes an essential step towards reducing the annotation budget of current approaches, as well as providing a more comprehensive understanding of the visual attention mechanism. Our code is available at: https://github.com/ashleylqx/WeakFixation.git.

Acquiring sufficient ground-truth supervision to train deep visual models has been a bottleneck over the years due to the data-hungry nature of deep learning. This is exacerbated in some structured prediction tasks, such as semantic segmentation, which require pixel-level annotations. This work addresses weakly supervised semantic segmentation (WSSS), with the goal of bridging the gap between image-level annotations and pixel-level segmentation. To achieve this, we propose, for the first time, a novel group-wise learning framework for WSSS. The framework explicitly encodes semantic dependencies in a group of images to discover rich semantic context for estimating more reliable pseudo ground-truths, which are subsequently employed to train more effective segmentation models. In particular, we solve the group-wise learning within a graph neural network (GNN), wherein input images are represented as graph nodes, and the underlying relations between a pair of images are characterized by graph edges. We then formulate semantic mining as an iterative reasoning process which propagates the common semantics shared by a group of images to enrich node representations. Moreover, in order to prevent the model from paying excessive attention to common semantics, we further propose a graph dropout layer to encourage the graph model to capture more accurate and complete object responses. With the above efforts, our model lays the foundation for more sophisticated and flexible group-wise semantic mining. We conduct comprehensive experiments on the popular PASCAL VOC 2012 and COCO benchmarks, and our model yields state-of-the-art performance. In addition, our model shows promising performance in weakly supervised object localization (WSOL) on the CUB-200-2011 dataset, demonstrating strong generalizability. Our code is available at: https://github.com/Lixy1997/Group-WSSS.

Performances of deep neural networks are prone to be degraded by label noise due to their powerful capability in fitting training data. Deeming low-loss instances as clean data is one of the most promising strategies in tackling label noise and has been widely adopted by state-of-the-art methods. However, prior works tend to drop high-loss instances directly, neglecting their valuable information. To address this issue, we propose an end-to-end framework named Co-LDL, which incorporates the low-loss sample selection strategy with label distribution learning. Specifically, we simultaneously train two deep neural networks and let them communicate useful knowledge by selecting low-loss and high-loss samples for each other. Low-loss samples are leveraged conventionally for updating network parameters. On the contrary, high-loss samples are trained in a label distribution learning manner to update network parameters and label distributions concurrently. Moreover, we propose a self-supervised module to further boost the model performance by enhancing the learned representations. Comprehensive experiments on both synthetic and real-world noisy datasets are provided to demonstrate the superiority of our Co-LDL method over state-of-the-art approaches in learning with noisy labels. The source code and models have been made available athttps://github.com/NUST-Machine-Intelligence-Laboratory/CoLDL.

Acquiring sufficient ground-truth supervision to train deep vi- sual models has been a bottleneck over the years due to the data-hungry nature of deep learning. This is exacerbated in some structured prediction tasks, such as semantic segmen- tation, which requires pixel-level annotations. This work ad- dresses weakly supervised semantic segmentation (WSSS), with the goal of bridging the gap between image-level anno- tations and pixel-level segmentation. We formulate WSSS as a novel group-wise learning task that explicitly models se- mantic dependencies in a group of images to estimate more reliable pseudo ground-truths, which can be used for training more accurate segmentation models. In particular, we devise a graph neural network (GNN) for group-wise semantic min- ing, wherein input images are represented as graph nodes, and the underlying relations between a pair of images are char- acterized by an efficient co-attention mechanism. Moreover, in order to prevent the model from paying excessive atten- tion to common semantics only, we further propose a graph dropout layer, encouraging the model to learn more accurate and complete object responses. The whole network is end-to- end trainable by iterative message passing, which propagates interaction cues over the images to progressively improve the performance. We conduct experiments on the popular PAS- CAL VOC 2012 and COCO benchmarks, and our model yields state-of-the-art performance. Our code is available at: https://github.com/Lixy1997/Group-WSSS.

Chest X-rays are an important and accessible clinical imaging tool for the detection of many thoracic diseases. Over the past decade, deep learning, with a focus on the convolutional neural network (CNN), has become the most powerful computer-aided diagnosis technology for improving disease identification performance. However, training an effective and robust deep CNN usually requires a large amount of data with high annotation quality. For chest X-ray imaging, annotating large-scale data requires professional domain knowledge and is time-consuming. Thus, existing public chest X-ray datasets usually adopt language pattern based methods to automatically mine labels from reports. However, this results in label uncertainty and inconsistency. In this paper, we propose many-to-one distribution learning (MODL) and K-nearest neighbor smoothing (KNNS) methods from two perspectives to improve a single model's disease identification performance, rather than focusing on an ensemble of models. MODL integrates multiple models to obtain a soft label distribution for optimizing the single target model, which can reduce the effects of original label uncertainty. Moreover, KNNS aims to enhance the robustness of the target model to provide consistent predictions on images with similar medical findings. Extensive experiments on the public NIH Chest X-ray and CheXpert datasets show that our model achieves consistent improvements over the state-of-the-art methods.

This paper addresses the task of unsupervised video multi-object segmentation. Current approaches follow a two-stage paradigm: 1) detect object proposals using pre-trained Mask R-CNN, and 2) conduct generic feature matching for temporal association using re-identification techniques. However, the generic features, widely used in both stages, are not reliable for characterizing unseen objects, leading to poor generalization. To address this, we introduce a novel approach for more accurate and efficient spatio-temporal segmentation. In particular, to address instance discrimination, we propose to combine foreground region estimation and instance grouping together in one network, and additionally introduce temporal guidance for segmenting each frame, enabling more accurate object discovery. For temporal association, we complement current video object segmentation architectures with a discriminative appearance model, capable of capturing more fine-grained target-specific information. Given object proposals from the instance discrimination network, three essential strategies are adopted to achieve accurate segmentation: 1) target-specific tracking using a memory-augmented appearance model; 2) target-agnostic verification to trace possible tracklets for the proposal; 3) adaptive memory updating using the verified segments. We evaluate the proposed approach on DAVIS17and YouTube-VIS, and the results demonstrate that it outperforms state-of-the-art methods both in segmentation accuracy and inference speed.

To address the challenging task of instance-aware human part parsing, a new bottom-up regime is proposed to learn category-level human semantic segmentation as well as multi-person pose estimation in a joint and end-to-end manner. It is a compact, efficient and powerful framework that exploits structural information over different human granularities and eases the difficulty of person partitioning. Specifically, a dense-to-sparse projection field, which allows explicitly associating dense human semantics with sparse keypoints, is learnt and progressively improved over the network feature pyramid for robustness. Then, the difficult pixel grouping problem is cast as an easier, multi-person joint assembling task. By formulating joint association as maximum-weight bipartite matching, a differentiable solution is developed to exploit projected gradient descent and Dykstra’s cyclic projection algorithm. This makes our method end-to-end trainable and allows back-propagating the grouping error to directly supervise multi-granularity human representation learning. This is distinguished from current bottom-up human parsers or pose estimators which require sophisticated post-processing or heuristic greedy algorithms. Experiments on three instance-aware human parsing datasets show that our model outperforms other bottom-up alternatives with much more efficient inference.

On existing public benchmarks, face forgery detection techniques have achieved great success. However, when used in multi-person videos, which often contain many people active in the scene with only a small subset having been manipulated, their performance remains far from being satisfactory. To take face forgery detection to a new level, we construct a novel large-scale dataset, called FFIW10K, which comprises 10,000 high-quality forgery videos, with an average of three human faces in each frame. The manipulation procedure is fully automatic, controlled by a domain-adversarial quality assessment network, making our dataset highly scalable with low human cost. In addition, we propose a novel algorithm to tackle the task of multi-person face forgery detection. Supervised by only video-level label, the algorithm explores multiple instance learning and learns to automatically attend to tampered faces. Our algorithm outperforms representative approaches for both forgery classification and localization on FFIW10K, and also shows high generalization ability on existing benchmarks. We hope that our dataset and study will help the community to explore this new field in more depth.

Current semantic segmentation methods focus only on mining "local" context, i.e., dependencies between pixels within individual images, by context-aggregation modules (e.g., dilated convolution, neural attention) or structure-aware optimization criteria (e.g., IoU-like loss). However, they ignore "global" context of the training data, i.e., rich semantic relations between pixels across different images. Inspired by recent advance in unsupervised contrastive representation learning, we propose a pixel-wise contrastive algorithm for semantic segmentation in the fully supervised setting. The core idea is to enforce pixel embeddings belonging to a same semantic class to be more similar than embeddings from different classes. It raises a pixel-wise metric learning paradigm for semantic segmentation, by explicitly exploring the structures of labeled pixels, which were rarely explored before. Our method can be effortlessly incorporated into existing segmentation frameworks without extra overhead during testing. We experimentally show that, with famous segmentation models (i.e., DeepLabV3, HRNet, OCR) and backbones (i.e., ResNet, HRNet), our method brings performance improvements across diverse datasets (i.e., Cityscapes, PASCAL-Context, COCO-Stuff, CamVid). We expect this work will encourage our community to rethink the current de facto training paradigm in semantic segmentation.

Automatic thoracic disease diagnosis is a rising research topic in the medical imaging community, with many potential applications. However, the inconsistent appearances and high complexities of various lesions in chest X-rays currently hinder the development of a reliable and robust intelligent diagnosis system. Attending to the high-probability abnormal regions and exploiting the priori of a related knowledge graph offers one promising route to addressing these issues. As such, in this paper, we propose two contrastive abnormal attention models and a dual-weighting graph convolution to improve the performance of thoracic multi-disease recognition. First, a left-right lung contrastive network is designed to learn intra-attentive abnormal features to better identify the most common thoracic diseases, whose lesions rarely appear in both sides symmetrically. Moreover, an inter-contrastive abnormal attention model aims to compare the query scan with multiple anchor scans without lesions to compute the abnormal attention map. Once the intra- and inter-contrastive attentions are weighted over the features, in addition to the basic visual spatial convolution, a chest radiology graph is constructed for dual-weighting graph reasoning. Extensive experiments on the public NIH ChestX-ray and CheXpert datasets show that our model achieves consistent improvements over the state-of-the-art methods both on thoracic disease identification and localization.

Labeling objects at the subordinate level typically requires expert knowledge, which is not always available from a random annotator. Accordingly, learning directly from web images for fine-grained visual classification (FGVC) has attracted broad attention. However, the existence of noise in web images is a huge obstacle for training robust deep neural networks. In this paper, we propose a novel approach to remove irrelevant samples from the real-world web images during training, and only utilize useful images for updating the networks. Thus, our network can alleviate the harmful effects caused by irrelevant noisy web images to achieve better performance. Extensive experiments on three commonly used fine-grained datasets demonstrate that our approach is much superior to state-of-the-art webly supervised methods. The data and source code of this work have been made anonymously available at: https://github.com/z337-408/WSNFGVC.

In this paper, we present a novel Motion-Attentive Transition Network (MATNet) for zero-shot video object segmentation, which provides a new way of leveraging motion information to reinforce spatio-temporal object representation. An asymmetric attention block, called Motion-Attentive Transition (MAT), is designed within a two-stream encoder, which transforms appearance features into motion-attentive representations at each convolutional stage. In this way, the encoder becomes deeply interleaved, allowing for closely hierarchical interactions between object motion and appearance. This is superior to the typical two-stream architecture, which treats motion and appearance separately in each stream and often suffers from overfitting to appearance information. Additionally, a bridge network is proposed to obtain a compact, discriminative and scale-sensitive representation for multi-level encoder features, which is further fed into a decoder to achieve segmentation results. Extensive experiments on three challenging public benchmarks (i.e., DAVIS-16, FBMS and Youtube-Objects) show that our model achieves compelling performance against the state-of-the-arts. Code is available at: https://github.com/tfzhou/MATNet.

Rapid progress has been witnessed for human-object interaction (HOI) recognition, but most existing models are confined to single-stage reasoning pipelines. Considering the intrinsic complexity of the task, we introduce a cascade architecture for a multi-stage, coarse-to-fine HOI understanding. At each stage, an instance localization network progressively refines HOI proposals and feeds them into an interaction recognition network. Each of the two networks is also connected to its predecessor at the previous stage, enabling cross-stage information propagation. The interaction recognition network has two crucial parts: a relation ranking module for high-quality HOI proposal selection and a triple-stream classifier for relation prediction. With our carefully-designed human-centric relation features, these two modules work collaboratively towards effective interaction understanding. Further beyond relation detection on a bounding-box level, we make ourframework flexible to perform fine-grained pixel-wise relation segmentation; this provides a new glimpse into better relation modeling. Our approach reached the 1st place in the ICCV2019 Person in Context Challenge, on both relation detection and segmentation tasks. It also shows promising results on V-COCO.

In this paper, we present a novel end-to-end learning neural network, i.e., MATNet, for zero-shot video object segmentation (ZVOS). Motivated by the human visual attention behavior, MATNet leverages motion cues as a bottom-up signal to guide the perception of object appearance. To achieve this, an asymmetric attention block, named Motion-Attentive Transition (MAT), is proposed within a two-stream encoder network to firstly identify moving regions and then attend appearance learning to capture the full extent of objects. Putting MATs in different convolutional layers, our encoder becomes deeply interleaved, allowing for close hierarchical interactions between object apperance and motion. Such a biologically-inspired design is proven to be superb to conventional two-stream structures, which treat motion and appearance independently in separate streams and often suffer severe overfitting to object appearance. Moreover, we introduce a bridge network to modulate multi-scale spatiotemporal features into more compact, discriminative and scale-sensitive representations, which are subsequently fed into a boundary-aware decoder network to produce accurate segmentation with crisp boundaries. We perform extensive quantitative and qualitative experiments on four challenging public benchmarks, i.e., DAVIS16, DAVIS17, FBMS and YouTube-Objects. Results show that our method achieves compelling performance against current state-of-the-art ZVOS methods. To further demonstrate the generalization ability of our spatiotemporal learning framework, we extend MATNet to another relevant task: dynamic visual attention prediction (DVAP). The experiments on two popular datasets (i.e., Hollywood-2 and UCF-Sports) further verify the superiority of our model. Our implementations have been made publicly available at https://github.com/tfzhou/MATNet.

This paper proposes a fully automatic pipeline to generate accurate object segment proposals in realistic videos. Our approach first detects generic object proposals for all video frames and then learns to rank them using a Convolutional Neural Networks (CNN) descriptor built on appearance and motion cues. The ambiguity of the proposal set can be reduced while the quality can be retained as highly as possible Next, high-scoring proposals are greedily tracked over the entire sequence into distinct tracklets. Observing that the proposal tracklet set at this stage is noisy and redundant, we perform a tracklet selection scheme to suppress the highly overlapped tracklets, and detect occlusions based on appearance and location information. Finally, we exploit holistic appearance cues for refinement of video segment proposals to obtain pixel-accurate segmentation. Our method is evaluated on two video segmentation datasets i.e. SegTrack v1 and FBMS-59 and achieves competitive results in comparison with other state-of-the-art methods.

This paper presents a novel discriminative, generative, and collaborative appearance model for robust object tracking. In contrast to existing methods, we use different appearance manifolds to represent the target in the discriminative and generative appearance models and propose a novel collaborative scheme to combine these two components. In particular: 1) for the discriminative component, we develop a graph regularized discriminant analysis (GRDA) algorithm that can find a projection to more effectively distinguish the target from the background; 2) for the generative component, we introduce a simple yet effective coding method for object representation. The method involves no optimization, and thus better efficiency can be achieved; and 3) for the collaborative model, we apply GRDA again to find a subspace for discriminating the likelihood features (generated from the discriminative and generative appearance models) and use the nearest neighbor criterion to determine the final likelihood. Besides, all the components are online updated so that our tracker can deal with appearance changes effectively. The experimental results over 23 challenging image sequences demonstrate that the proposed algorithm achieves better performance compared with other state-of-the-art methods.

Multiple Instance Learning (MIL) recently provides an appealing way to alleviate the drifting problem in visual tracking. Following the tracking-by-detection framework, an online MILBoost approach is developed that sequentially chooses weak classifiers by maximizing the bag likelihood. In this paper, we extend this idea towards incorporating the instance significance estimation into the online MILBoost framework. First, instead of treating all instances equally, with each instance we associate a significance-coefficient that represents its contribution to the bag likelihood. The coefficients are estimated by a Bayesian formula that jointly considers the predictions from multiple randomized MILBoost classifiers. Next, we incorporate the estimates within a new boosting procedure for more effectively selection of weak classifiers. Experiments with challenging public datasets show that the proposed method outperforms both existing MIL based and boosting based trackers.

We present an approach for unsupervised object segmentation in unconstrained videos. Driven by the latest progress in this field, we argue that segmentation performance can be largely improved by aggregating the results generated by state-of-the-art algorithms. Initially, objects in individual frames are estimated through a per-frame aggregation procedure using majority voting. While this can predict relatively accurate object location, the initial estimation fails to cover the parts that are wrongly labeled by more than half of the algorithms. To address this, we build a holistic appearance model using non-local appearance cues by linear regression. Then, we integrate the appearance priors and spatio-temporal information into an energy minimization framework to refine the initial estimation. We evaluate our method on challenging benchmark videos and demonstrate that it outperforms state-of-the-art algorithms.

Linear Discriminant Analysis (LDA) is a powerful technology for supervised dimensionality reduction, however, it only captures the extrinsic (or global) structure in the data and fails to discover the intrinsic structure of the data manifold. In this paper, we develop a new linear supervised dimensionality reduction method, called Graph Regularized Discriminant Analysis(GRDA), which respects both extrinsic and intrinsic structure in the data. In particular, a regularization term, incorporating the manifold structure, is introduced into the objective function of LDA. The formulation allows us to achieve a more discriminative subspace by simultaneously considering the graph preserving and the global LDA criteria. We then apply the proposed GRDA algorithm to face recognition by exploiting the local dissimilarity of face images in different classes. Experimental results clearly show that the proposed GRDA method outperforms many state-of-the-art face recognition algorithms.

Stochastic sampling based trackers have shown good performance for abrupt motion tracking so that they have gained popularity in recent years. However, the existing methods tend to explore the whole state space uniformly with an inefficiency preliminary sampling phase. In this paper, we propose a nearest neighbor field(NNF) driven stochastic sampling framework for abrupt motion tracking in which NNF provides us promising regions the target may exist, and thus can help to explore the state space more effectively. Our approach firstly computes NNF to determine the promising regions; subsequently, we adopt Smoothing Stochastic Approximate Monte Carlo(SSAMC) sampling scheme to accurately localize the target. SSAMC is robust to handle the noises in NNF by propagating a sample's information to its neighboring regions. Finally, we refine the result with sparse representation based template matching technique. The experimental results on challenging sequences show that our tracker outperforms other related methods by better accuracy and higher robustness.