Wei Feng

Xiaoguang Li, Qing Guo, Di Lin et al.

Although achieving significant progress, existing deep generative inpainting methods are far from real-world applications due to the low generalization across different scenes. As a result, the generated images usually contain artifacts or the filled pixels differ greatly from the ground truth. Image-level predictive filtering is a widely used image restoration technique, predicting suitable kernels adaptively according to different input scenes. Inspired by this inherent advantage, we explore the possibility of addressing image inpainting as a filtering task. To this end, we first study the advantages and challenges of image-level predictive filtering for image inpainting: the method can preserve local structures and avoid artifacts but fails to fill large missing areas. Then, we propose semantic filtering by conducting filtering on the deep feature level, which fills the missing semantic information but fails to recover the details. To address the issues while adopting the respective advantages, we propose a novel filtering technique, i.e., Multilevel Interactive Siamese Filtering (MISF), which contains two branches: kernel prediction branch (KPB) and semantic & image filtering branch (SIFB). These two branches are interactively linked: SIFB provides multi-level features for KPB while KPB predicts dynamic kernels for SIFB. As a result, the final method takes the advantage of effective semantic & image-level filling for high-fidelity inpainting. We validate our method on three challenging datasets, i.e., Dunhuang, Places2, and CelebA. Our method outperforms state-of-the-art baselines on four metrics, i.e., L1, PSNR, SSIM, and LPIPS. Please try the released code and model at https://github.com/tsingqguo/misf.

Lianyu Hu, Liqing Gao, Zekang Liu et al.

Raw videos have been proven to own considerable feature redundancy where in many cases only a portion of frames can already meet the requirements for accurate recognition. In this paper, we are interested in whether such redundancy can be effectively leveraged to facilitate efficient inference in continuous sign language recognition (CSLR). We propose a novel adaptive model (AdaBrowse) to dynamically select a most informative subsequence from input video sequences by modelling this problem as a sequential decision task. In specific, we first utilize a lightweight network to quickly scan input videos to extract coarse features. Then these features are fed into a policy network to intelligently select a subsequence to process. The corresponding subsequence is finally inferred by a normal CSLR model for sentence prediction. As only a portion of frames are processed in this procedure, the total computations can be considerably saved. Besides temporal redundancy, we are also interested in whether the inherent spatial redundancy can be seamlessly integrated together to achieve further efficiency, i.e., dynamically selecting a lowest input resolution for each sample, whose model is referred to as AdaBrowse+. Extensive experimental results on four large-scale CSLR datasets, i.e., PHOENIX14, PHOENIX14-T, CSL-Daily and CSL, demonstrate the effectiveness of AdaBrowse and AdaBrowse+ by achieving comparable accuracy with state-of-the-art methods with 1.44$\times$ throughput and 2.12$\times$ fewer FLOPs. Comparisons with other commonly-used 2D CNNs and adaptive efficient methods verify the effectiveness of AdaBrowse. Code is available at \url{https://github.com/hulianyuyy/AdaBrowse}.

Zhijin Ge, Fanhua Shang, Hongying Liu et al.

Deep neural networks are vulnerable to adversarial examples crafted by applying human-imperceptible perturbations on clean inputs. Although many attack methods can achieve high success rates in the white-box setting, they also exhibit weak transferability in the black-box setting. Recently, various methods have been proposed to improve adversarial transferability, in which the input transformation is one of the most effective methods. In this work, we notice that existing input transformation-based works mainly adopt the transformed data in the same domain for augmentation. Inspired by domain generalization, we aim to further improve the transferability using the data augmented from different domains. Specifically, a style transfer network can alter the distribution of low-level visual features in an image while preserving semantic content for humans. Hence, we propose a novel attack method named Style Transfer Method (STM) that utilizes a proposed arbitrary style transfer network to transform the images into different domains. To avoid inconsistent semantic information of stylized images for the classification network, we fine-tune the style transfer network and mix up the generated images added by random noise with the original images to maintain semantic consistency and boost input diversity. Extensive experimental results on the ImageNet-compatible dataset show that our proposed method can significantly improve the adversarial transferability on either normally trained models or adversarially trained models than state-of-the-art input transformation-based attacks. Code is available at: https://github.com/Zhijin-Ge/STM.

Lianyu Hu, Liqing Gao, Zekang liu et al.

Hand and face play an important role in expressing sign language. Their features are usually especially leveraged to improve system performance. However, to effectively extract visual representations and capture trajectories for hands and face, previous methods always come at high computations with increased training complexity. They usually employ extra heavy pose-estimation networks to locate human body keypoints or rely on additional pre-extracted heatmaps for supervision. To relieve this problem, we propose a self-emphasizing network (SEN) to emphasize informative spatial regions in a self-motivated way, with few extra computations and without additional expensive supervision. Specifically, SEN first employs a lightweight subnetwork to incorporate local spatial-temporal features to identify informative regions, and then dynamically augment original features via attention maps. It's also observed that not all frames contribute equally to recognition. We present a temporal self-emphasizing module to adaptively emphasize those discriminative frames and suppress redundant ones. A comprehensive comparison with previous methods equipped with hand and face features demonstrates the superiority of our method, even though they always require huge computations and rely on expensive extra supervision. Remarkably, with few extra computations, SEN achieves new state-of-the-art accuracy on four large-scale datasets, PHOENIX14, PHOENIX14-T, CSL-Daily, and CSL. Visualizations verify the effects of SEN on emphasizing informative spatial and temporal features. Code is available at https://github.com/hulianyuyy/SEN_CSLR

Kaile Du, Fan Lyu, Fuyuan Hu et al.

The Lifelong Multi-Label (LML) image recognition builds an online class-incremental classifier in a sequential multi-label image recognition data stream. The key challenges of LML image recognition are the construction of label relationships on Partial Labels of training data and the Catastrophic Forgetting on old classes, resulting in poor generalization. To solve the problems, the study proposes an Augmented Graph Convolutional Network (AGCN) model that can construct the label relationships across the sequential recognition tasks and sustain the catastrophic forgetting. First, we build an Augmented Correlation Matrix (ACM) across all seen classes, where the intra-task relationships derive from the hard label statistics while the inter-task relationships leverage both hard and soft labels from data and a constructed expert network. Then, based on the ACM, the proposed AGCN captures label dependencies with dynamic augmented structure and yields effective class representations. Last, to suppress the forgetting of label dependencies across old tasks, we propose a relationship-preserving loss as a constraint to the construction of label relationships. The proposed method is evaluated using two multi-label image benchmarks and the experimental results show that the proposed method is effective for LML image recognition and can build convincing correlation across tasks even if the labels of previous tasks are missing. Our code is available at https://github.com/Kaile-Du/AGCN.

Lie Ju, Yicheng Wu, Wei Feng et al.

Semi-supervised learning (SSL) has attracted much attention since it reduces the expensive costs of collecting adequate well-labeled training data, especially for deep learning methods. However, traditional SSL is built upon an assumption that labeled and unlabeled data should be from the same distribution \textit{e.g.,} classes and domains. However, in practical scenarios, unlabeled data would be from unseen classes or unseen domains, and it is still challenging to exploit them by existing SSL methods. Therefore, in this paper, we proposed a unified framework to leverage these unseen unlabeled data for open-scenario semi-supervised medical image classification. We first design a novel scoring mechanism, called dual-path outliers estimation, to identify samples from unseen classes. Meanwhile, to extract unseen-domain samples, we then apply an effective variational autoencoder (VAE) pre-training. After that, we conduct domain adaptation to fully exploit the value of the detected unseen-domain samples to boost semi-supervised training. We evaluated our proposed framework on dermatology and ophthalmology tasks. Extensive experiments demonstrate our model can achieve superior classification performance in various medical SSL scenarios. The code implementations are accessible at: https://github.com/PyJulie/USSL4MIC.

Ming Hu, Lin Wang, Siyuan Yan et al.

The application of deep learning to nursing procedure activity understanding has the potential to greatly enhance the quality and safety of nurse-patient interactions. By utilizing the technique, we can facilitate training and education, improve quality control, and enable operational compliance monitoring. However, the development of automatic recognition systems in this field is currently hindered by the scarcity of appropriately labeled datasets. The existing video datasets pose several limitations: 1) these datasets are small-scale in size to support comprehensive investigations of nursing activity; 2) they primarily focus on single procedures, lacking expert-level annotations for various nursing procedures and action steps; and 3) they lack temporally localized annotations, which prevents the effective localization of targeted actions within longer video sequences. To mitigate these limitations, we propose NurViD, a large video dataset with expert-level annotation for nursing procedure activity understanding. NurViD consists of over 1.5k videos totaling 144 hours, making it approximately four times longer than the existing largest nursing activity datasets. Notably, it encompasses 51 distinct nursing procedures and 177 action steps, providing a much more comprehensive coverage compared to existing datasets that primarily focus on limited procedures. To evaluate the efficacy of current deep learning methods on nursing activity understanding, we establish three benchmarks on NurViD: procedure recognition on untrimmed videos, procedure and action recognition on trimmed videos, and action detection. Our benchmark and code will be available at \url{https://github.com/minghu0830/NurViD-benchmark}.

Fengheng Li, An Liu, Wei Feng et al.

Poster layout is a crucial aspect of poster design. Prior methods primarily focus on the correlation between visual content and graphic elements. However, a pleasant layout should also consider the relationship between visual and textual contents and the relationship between elements. In this study, we introduce a relation-aware diffusion model for poster layout generation that incorporates these two relationships in the generation process. Firstly, we devise a visual-textual relation-aware module that aligns the visual and textual representations across modalities, thereby enhancing the layout's efficacy in conveying textual information. Subsequently, we propose a geometry relation-aware module that learns the geometry relationship between elements by comprehensively considering contextual information. Additionally, the proposed method can generate diverse layouts based on user constraints. To advance research in this field, we have constructed a poster layout dataset named CGL-Dataset V2. Our proposed method outperforms state-of-the-art methods on CGL-Dataset V2. The data and code will be available at https://github.com/liuan0803/RADM.

Linkai Liu, Wei Feng, Xi Zhao et al.

Creative Generation (CG) leverages generative models to automatically produce advertising content that highlights product features, and it has been a significant focus of recent research. However, while CG has advanced considerably, most efforts have concentrated on generating advertising text and images, leaving Creative Video Generation (CVG) relatively underexplored. This gap is largely due to two major challenges faced by Text-to-Video (T2V) models: (a) \textbf{ambiguous semantic alignment}, where models struggle to accurately correlate product selling points with creative video content, and (b) \textbf{inadequate motion adaptability}, resulting in unrealistic movements and distortions. To address these challenges, we develop a comprehensive Advertising Creative Knowledge Base (ACKB) as a foundational resource and propose a knowledge-driven approach (KD-CVG) to overcome the knowledge limitations of existing models. KD-CVG consists of two primary modules: Semantic-Aware Retrieval (SAR) and Multimodal Knowledge Reference (MKR). SAR utilizes the semantic awareness of graph attention networks and reinforcement learning feedback to enhance the model's comprehension of the connections between selling points and creative videos. Building on this, MKR incorporates semantic and motion priors into the T2V model to address existing knowledge gaps. Extensive experiments have demonstrated KD-CVG's superior performance in achieving semantic alignment and motion adaptability, validating its effectiveness over other state-of-the-art methods. The code and dataset will be open source at https://kdcvg.github.io/KDCVG/.

Xiaoguang Li, Qing Guo, Rabab Abdelfattah et al.

Fully-supervised shadow removal methods achieve the best restoration qualities on public datasets but still generate some shadow remnants. One of the reasons is the lack of large-scale shadow & shadow-free image pairs. Unsupervised methods can alleviate the issue but their restoration qualities are much lower than those of fully-supervised methods. In this work, we find that pretraining shadow removal networks on the image inpainting dataset can reduce the shadow remnants significantly: a naive encoder-decoder network gets competitive restoration quality w.r.t. the state-of-the-art methods via only 10% shadow & shadow-free image pairs. After analyzing networks with/without inpainting pre-training via the information stored in the weight (IIW), we find that inpainting pretraining improves restoration quality in non-shadow regions and enhances the generalization ability of networks significantly. Additionally, shadow removal fine-tuning enables networks to fill in the details of shadow regions. Inspired by these observations we formulate shadow removal as an adaptive fusion task that takes advantage of both shadow removal and image inpainting. Specifically, we develop an adaptive fusion network consisting of two encoders, an adaptive fusion block, and a decoder. The two encoders are responsible for extracting the feature from the shadow image and the shadow-masked image respectively. The adaptive fusion block is responsible for combining these features in an adaptive manner. Finally, the decoder converts the adaptive fused features to the desired shadow-free result. The extensive experiments show that our method empowered with inpainting outperforms all state-of-the-art methods.

Xi Zhao, Wei Feng, Zheng Zhang et al.

Recently, segmentation-based methods are quite popular in scene text detection, which mainly contain two steps: text kernel segmentation and expansion. However, the segmentation process only considers each pixel independently, and the expansion process is difficult to achieve a favorable accuracy-speed trade-off. In this paper, we propose a Context-aware and Boundary-guided Network (CBN) to tackle these problems. In CBN, a basic text detector is firstly used to predict initial segmentation results. Then, we propose a context-aware module to enhance text kernel feature representations, which considers both global and local contexts. Finally, we introduce a boundary-guided module to expand enhanced text kernels adaptively with only the pixels on the contours, which not only obtains accurate text boundaries but also keeps high speed, especially on high-resolution output maps. In particular, with a lightweight backbone, the basic detector equipped with our proposed CBN achieves state-of-the-art results on several popular benchmarks, and our proposed CBN can be plugged into several segmentation-based methods. Code is available at https://github.com/XiiZhao/cbn.pytorch.

Wei Feng, Lie Ju, Lin Wang et al. · ibm-research

Existing deep learning models have achieved promising performance in recognizing skin diseases from dermoscopic images. However, these models can only recognize samples from predefined categories, when they are deployed in the clinic, data from new unknown categories are constantly emerging. Therefore, it is crucial to automatically discover and identify new semantic categories from new data. In this paper, we propose a new novel class discovery framework for automatically discovering new semantic classes from dermoscopy image datasets based on the knowledge of known classes. Specifically, we first use contrastive learning to learn a robust and unbiased feature representation based on all data from known and unknown categories. We then propose an uncertainty-aware multi-view cross pseudo-supervision strategy, which is trained jointly on all categories of data using pseudo labels generated by a self-labeling strategy. Finally, we further refine the pseudo label by aggregating neighborhood information through local sample similarity to improve the clustering performance of the model for unknown categories. We conducted extensive experiments on the dermatology dataset ISIC 2019, and the experimental results show that our approach can effectively leverage knowledge from known categories to discover new semantic categories. We also further validated the effectiveness of the different modules through extensive ablation experiments. Our code will be released soon.

Yuhao Liu, Qing Guo, Lan Fu et al.

Existing deep learning-based shadow removal methods still produce images with shadow remnants. These shadow remnants typically exist in homogeneous regions with low-intensity values, making them untraceable in the existing image-to-image mapping paradigm. We observe that shadows mainly degrade images at the image-structure level (in which humans perceive object shapes and continuous colors). Hence, in this paper, we propose to remove shadows at the image structure level. Based on this idea, we propose a novel structure-informed shadow removal network (StructNet) to leverage the image-structure information to address the shadow remnant problem. Specifically, StructNet first reconstructs the structure information of the input image without shadows and then uses the restored shadow-free structure prior to guiding the image-level shadow removal. StructNet contains two main novel modules: (1) a mask-guided shadow-free extraction (MSFE) module to extract image structural features in a non-shadow-to-shadow directional manner, and (2) a multi-scale feature & residual aggregation (MFRA) module to leverage the shadow-free structure information to regularize feature consistency. In addition, we also propose to extend StructNet to exploit multi-level structure information (MStructNet), to further boost the shadow removal performance with minimum computational overheads. Extensive experiments on three shadow removal benchmarks demonstrate that our method outperforms existing shadow removal methods, and our StructNet can be integrated with existing methods to improve them further.

Qing Sun, Fan Lyu, Fanhua Shang et al.

Continual Learning (CL) sequentially learns new tasks like human beings, with the goal to achieve better Stability (S, remembering past tasks) and Plasticity (P, adapting to new tasks). Due to the fact that past training data is not available, it is valuable to explore the influence difference on S and P among training examples, which may improve the learning pattern towards better SP. Inspired by Influence Function (IF), we first study example influence via adding perturbation to example weight and computing the influence derivation. To avoid the storage and calculation burden of Hessian inverse in neural networks, we propose a simple yet effective MetaSP algorithm to simulate the two key steps in the computation of IF and obtain the S- and P-aware example influence. Moreover, we propose to fuse two kinds of example influence by solving a dual-objective optimization problem, and obtain a fused influence towards SP Pareto optimality. The fused influence can be used to control the update of model and optimize the storage of rehearsal. Empirical results show that our algorithm significantly outperforms state-of-the-art methods on both task- and class-incremental benchmark CL datasets.

Wei Feng, Lin Wang, Lie Ju et al.

Existing unsupervised domain adaptation methods based on adversarial learning have achieved good performance in several medical imaging tasks. However, these methods focus only on global distribution adaptation and ignore distribution constraints at the category level, which would lead to sub-optimal adaptation performance. This paper presents an unsupervised domain adaptation framework based on category-level regularization that regularizes the category distribution from three perspectives. Specifically, for inter-domain category regularization, an adaptive prototype alignment module is proposed to align feature prototypes of the same category in the source and target domains. In addition, for intra-domain category regularization, we tailored a regularization technique for the source and target domains, respectively. In the source domain, a prototype-guided discriminative loss is proposed to learn more discriminative feature representations by enforcing intra-class compactness and inter-class separability, and as a complement to traditional supervised loss. In the target domain, an augmented consistency category regularization loss is proposed to force the model to produce consistent predictions for augmented/unaugmented target images, which encourages semantically similar regions to be given the same label. Extensive experiments on two publicly fundus datasets show that the proposed approach significantly outperforms other state-of-the-art comparison algorithms.

Fenghao Zhu, Xinquan Wang, Siming Jiang et al.

The emergence of sixth-generation and beyond communication systems is expected to fundamentally transform digital experiences through introducing unparalleled levels of intelligence, efficiency, and connectivity. A promising technology poised to enable this revolutionary vision is a wireless large AI model (WLAM), characterized by its exceptional capabilities in data processing, inference, and decision-making. In light of these remarkable capabilities, this paper provides a comprehensive survey of WLAM, explaining its fundamental principles, diverse applications, critical challenges, and future research opportunities. We begin by introducing the background of WLAM and analyzing the key synergies with wireless networks, emphasizing the mutual benefits. Subsequently, we explore the foundational characteristics of WLAM, delving into their unique relevance in wireless environments. Then, the role of WLAM in optimizing wireless communication systems across various use cases and the reciprocal benefits are systematically investigated. Furthermore, we discuss the integration of WLAM with emerging technologies, highlighting their potential to enable transformative capabilities and breakthroughs in wireless communication. Finally, we thoroughly examine the high-level challenges and discuss pivotal future research directions.

Jiahao Fan, Yuxin Qin, Wei Feng et al.

Product posters blend striking visuals with informative text to highlight the product and capture customer attention. However, crafting appealing posters and manually optimizing them based on online performance is laborious and resource-consuming. To address this, we introduce AutoPP, an automated pipeline for product poster generation and optimization that eliminates the need for human intervention. Specifically, the generator, relying solely on basic product information, first uses a unified design module to integrate the three key elements of a poster (background, text, and layout) into a cohesive output. Then, an element rendering module encodes these elements into condition tokens, efficiently and controllably generating the product poster. Based on the generated poster, the optimizer enhances its Click-Through Rate (CTR) by leveraging online feedback. It systematically replaces elements to gather fine-grained CTR comparisons and utilizes Isolated Direct Preference Optimization (IDPO) to attribute CTR gains to isolated elements. Our work is supported by AutoPP1M, the largest dataset specifically designed for product poster generation and optimization, which contains one million high-quality posters and feedback collected from over one million users. Experiments demonstrate that AutoPP achieves state-of-the-art results in both offline and online settings. Our code and dataset are publicly available at: https://github.com/JD-GenX/AutoPP

Jiacheng Li, Ruize Han, Haomin Yan et al.

Human group detection, which splits crowd of people into groups, is an important step for video-based human social activity analysis. The core of human group detection is the human social relation representation and division.In this paper, we propose a new two-stage multi-head framework for human group detection. In the first stage, we propose a human behavior simulator head to learn the social relation feature embedding, which is self-supervisely trained by leveraging the socially grounded multi-person behavior relationship. In the second stage, based on the social relation embedding, we develop a self-attention inspired network for human group detection. Remarkable performance on two state-of-the-art large-scale benchmarks, i.e., PANDA and JRDB-Group, verifies the effectiveness of the proposed framework. Benefiting from the self-supervised social relation embedding, our method can provide promising results with very few (labeled) training data. We will release the source code to the public.

Lianyu Hu, Liqing Gao, Zekang Liu et al.

Human body trajectories are a salient cue to identify actions in the video. Such body trajectories are mainly conveyed by hands and face across consecutive frames in sign language. However, current methods in continuous sign language recognition (CSLR) usually process frames independently, thus failing to capture cross-frame trajectories to effectively identify a sign. To handle this limitation, we propose correlation network (CorrNet) to explicitly capture and leverage body trajectories across frames to identify signs. In specific, a correlation module is first proposed to dynamically compute correlation maps between the current frame and adjacent frames to identify trajectories of all spatial patches. An identification module is then presented to dynamically emphasize the body trajectories within these correlation maps. As a result, the generated features are able to gain an overview of local temporal movements to identify a sign. Thanks to its special attention on body trajectories, CorrNet achieves new state-of-the-art accuracy on four large-scale datasets, i.e., PHOENIX14, PHOENIX14-T, CSL-Daily, and CSL. A comprehensive comparison with previous spatial-temporal reasoning methods verifies the effectiveness of CorrNet. Visualizations demonstrate the effects of CorrNet on emphasizing human body trajectories across adjacent frames.

Zekun Qian, Ruize Han, Wei Feng et al.

We tackle a new problem of multi-view camera and subject registration in the bird's eye view (BEV) without pre-given camera calibration. This is a very challenging problem since its only input is several RGB images from different first-person views (FPVs) for a multi-person scene, without the BEV image and the calibration of the FPVs, while the output is a unified plane with the localization and orientation of both the subjects and cameras in a BEV. We propose an end-to-end framework solving this problem, whose main idea can be divided into following parts: i) creating a view-transform subject detection module to transform the FPV to a virtual BEV including localization and orientation of each pedestrian, ii) deriving a geometric transformation based method to estimate camera localization and view direction, i.e., the camera registration in a unified BEV, iii) making use of spatial and appearance information to aggregate the subjects into the unified BEV. We collect a new large-scale synthetic dataset with rich annotations for evaluation. The experimental results show the remarkable effectiveness of our proposed method.

Kaile Du, Fan Lyu, Linyan Li et al.

Multi-Label Continual Learning (MLCL) builds a class-incremental framework in a sequential multi-label image recognition data stream. The critical challenges of MLCL are the construction of label relationships on past-missing and future-missing partial labels of training data and the catastrophic forgetting on old classes, resulting in poor generalization. To solve the problems, the study proposes an Augmented Graph Convolutional Network (AGCN++) that can construct the cross-task label relationships in MLCL and sustain catastrophic forgetting. First, we build an Augmented Correlation Matrix (ACM) across all seen classes, where the intra-task relationships derive from the hard label statistics. In contrast, the inter-task relationships leverage hard and soft labels from data and a constructed expert network. Then, we propose a novel partial label encoder (PLE) for MLCL, which can extract dynamic class representation for each partial label image as graph nodes and help generate soft labels to create a more convincing ACM and suppress forgetting. Last, to suppress the forgetting of label dependencies across old tasks, we propose a relationship-preserving constrainter to construct label relationships. The inter-class topology can be augmented automatically, which also yields effective class representations. The proposed method is evaluated using two multi-label image benchmarks. The experimental results show that the proposed way is effective for MLCL image recognition and can build convincing correlations across tasks even if the labels of previous tasks are missing.

Lianyu Hu, Liqing Gao, Zekang Liu et al.

Pooling methods are necessities for modern neural networks for increasing receptive fields and lowering down computational costs. However, commonly used hand-crafted pooling approaches, e.g., max pooling and average pooling, may not well preserve discriminative features. While many researchers have elaborately designed various pooling variants in spatial domain to handle these limitations with much progress, the temporal aspect is rarely visited where directly applying hand-crafted methods or these specialized spatial variants may not be optimal. In this paper, we derive temporal lift pooling (TLP) from the Lifting Scheme in signal processing to intelligently downsample features of different temporal hierarchies. The Lifting Scheme factorizes input signals into various sub-bands with different frequency, which can be viewed as different temporal movement patterns. Our TLP is a three-stage procedure, which performs signal decomposition, component weighting and information fusion to generate a refined downsized feature map. We select a typical temporal task with long sequences, i.e. continuous sign language recognition (CSLR), as our testbed to verify the effectiveness of TLP. Experiments on two large-scale datasets show TLP outperforms hand-crafted methods and specialized spatial variants by a large margin (1.5%) with similar computational overhead. As a robust feature extractor, TLP exhibits great generalizability upon multiple backbones on various datasets and achieves new state-of-the-art results on two large-scale CSLR datasets. Visualizations further demonstrate the mechanism of TLP in correcting gloss borders. Code is released.

Yupeng Zhang, Ruize Han, Wei Feng et al.

Object detection is an important task in computer vision, which aims to detect the objects of interest. through the given category list or query images. In this work, we propose a new problem of language-visual-complementary open-set object detection (LV-OSD), i.e., using the flexible text-based and/or image-based prompts to specify the desired object categories. This setting is more common and practical in real-world applications. For this purpose, we design a dual-branch detection framework, LVDor, which can simultaneously accept both text and image prompts. Specifically, we first build the Multi-modal Prompts (MPr) containing various text descriptions and image samples for each category. Subsequently, to bridge the semantic gap among the input image, text prompts, and image prompts, we design a Target-guided Prompt Dynamic Weighting (TPDW) module. Guided by the prior information of the target image, this module dynamically produces the text and image prompts that best align with the target semantics, achieving precise alignment and effectively reducing the discrepancy between the two modalities, thereby accommodating the LV-OSD setting. We also propose a simple Prompt Random Masking (PRM) mechanism during training to simulate the arbitrary combination of text and/or image prompts in testing. Extensive experimental results verify our problem formulation's reasonability and our method's effectiveness. Prompts and code will be released publicly.

Shuo Lu, Haohan Wang, Wei Feng et al.

Advertising image generation has increasingly focused on online metrics like Click-Through Rate (CTR), yet existing approaches adopt a ``one-size-fits-all" strategy that optimizes for overall CTR while neglecting preference diversity among user groups. This leads to suboptimal performance for specific groups, limiting targeted marketing effectiveness. To bridge this gap, we present \textit{One Size, Many Fits} (OSMF), a unified framework that aligns diverse group-wise click preferences in large-scale advertising image generation. OSMF begins with product-aware adaptive grouping, which dynamically organizes users based on their attributes and product characteristics, representing each group with rich collective preference features. Building on these groups, preference-conditioned image generation employs a Group-aware Multimodal Large Language Model (G-MLLM) to generate tailored images for each group. The G-MLLM is pre-trained to simultaneously comprehend group features and generate advertising images. Subsequently, we fine-tune the G-MLLM using our proposed Group-DPO for group-wise preference alignment, which effectively enhances each group's CTR on the generated images. To further advance this field, we introduce the Grouped Advertising Image Preference Dataset (GAIP), the first large-scale public dataset of group-wise image preferences, including around 600K groups built from 40M users. Extensive experiments demonstrate that our framework achieves the state-of-the-art performance in both offline and online settings. Our code and datasets will be released at https://github.com/JD-GenX/OSMF.

Ruize Han, Haomin Yan, Jiacheng Li et al.

To obtain a more comprehensive activity understanding for a crowded scene, in this paper, we propose a new problem of panoramic human activity recognition (PAR), which aims to simultaneous achieve the individual action, social group activity, and global activity recognition. This is a challenging yet practical problem in real-world applications. For this problem, we develop a novel hierarchical graph neural network to progressively represent and model the multi-granularity human activities and mutual social relations for a crowd of people. We further build a benchmark to evaluate the proposed method and other existing related methods. Experimental results verify the rationality of the proposed PAR problem, the effectiveness of our method and the usefulness of the benchmark. We will release the source code and benchmark to the public for promoting the study on this problem.

Yize Liu, Siyuan Yan, Ming Hu et al.

Dermatological diagnosis requires integrating fine-grained visual perception with expert clinical knowledge. Although Multimodal Large Language Models (MLLMs) facilitate interactive medical image analysis, their application in dermatology is hindered by insufficient domain-specific grounding and hallucinations. To address these issues, we propose DermAgent, a collaborative multi-tool agent that orchestrates seven specialized vision and language modules within a Plan-Execute-Reflect framework. DermAgent delivers stepwise, traceable diagnostic reasoning through three core components. First, it employs complementary visual perception tools for comprehensive morphological description, dermoscopic concept annotation, and disease diagnosis. Second, to overcome the lack of domain prior, a dual-modality retrieval module anchors every prediction in external evidence by cross-referencing 413,210 diagnosed image cases and 3,199 clinical guideline chunks. To further mitigate hallucinations, a deterministic critic module conducts strict post-hoc auditing via confidence, coverage, and conflict gates, automatically detecting inter-source disagreements to trigger targeted self-correction. Extensive experiments on five dermatology benchmarks demonstrate that DermAgent consistently outperforms state-of-the-art MLLMs and medical agent baselines across zero-shot fine-grained disease diagnosis, concept annotation, and clinical captioning tasks, exceeding GPT-4o by 17.6% in skin disease diagnostic accuracy and 3.15% in captioning ROUGE-L. Our code is available at https://github.com/YizeezLiu/DermAgent.

Lin Wang, Xiufen Ye, Donghao Zhang et al.

Usually, lesions are not isolated but are associated with the surrounding tissues. For example, the growth of a tumour can depend on or infiltrate into the surrounding tissues. Due to the pathological nature of the lesions, it is challenging to distinguish their boundaries in medical imaging. However, these uncertain regions may contain diagnostic information. Therefore, the simple binarization of lesions by traditional binary segmentation can result in the loss of diagnostic information. In this work, we introduce the image matting into the 3D scenes and use the alpha matte, i.e., a soft mask, to describe lesions in a 3D medical image. The traditional soft mask acted as a training trick to compensate for the easily mislabelled or under-labelled ambiguous regions. In contrast, 3D matting uses soft segmentation to characterize the uncertain regions more finely, which means that it retains more structural information for subsequent diagnosis and treatment. The current study of image matting methods in 3D is limited. To address this issue, we conduct a comprehensive study of 3D matting, including both traditional and deep-learning-based methods. We adapt four state-of-the-art 2D image matting algorithms to 3D scenes and further customize the methods for CT images to calibrate the alpha matte with the radiodensity. Moreover, we propose the first end-to-end deep 3D matting network and implement a solid 3D medical image matting benchmark. Its efficient counterparts are also proposed to achieve a good performance-computation balance. Furthermore, there is no high-quality annotated dataset related to 3D matting, slowing down the development of data-driven deep-learning-based methods. To address this issue, we construct the first 3D medical matting dataset. The validity of the dataset was verified through clinicians' assessments and downstream experiments.

Yexing Xu, Wei Feng, Shen Zhang et al.

Generating realistic and user-preferred advertisements is a key challenge in e-commerce. Existing approaches utilize multiple independent models driven by click-through-rate (CTR) to controllably create attractive image or text advertisements. However, their pipelines lack cross-modal perception and rely on CTR that only reflects average preferences. Therefore, we explore jointly generating personalized image-text advertisements from historical click behaviors. We first design a Unified Advertisement Generative model (Uni-AdGen) that employs a single autoregressive framework to produce both advertising images and texts. By incorporating a foreground perception module and instruction tuning, Uni-AdGen enhances the realism of the generated content. To further personalize advertisements, we equip Uni-AdGen with a coarse-to-fine preference understanding module that effectively captures user interests from noisy multimodal historical behaviors to drive personalized generation. Additionally, we construct the first large-scale Personalized Advertising image-text dataset (PAd1M) and introduce a Product Background Similarity (PBS) metric to facilitate training and evaluation. Extensive experiments show that our method outperforms baselines in general and personalized advertisement generation. Our project is available at https://github.com/JD-GenX/Uni-AdGen.

Ruize Han, Wei Feng, Qing Guo et al.

Visual object tracking is an important task in computer vision, which has many real-world applications, e.g., video surveillance, visual navigation. Visual object tracking also has many challenges, e.g., object occlusion and deformation. To solve above problems and track the target accurately and efficiently, many tracking algorithms have emerged in recent years. This paper presents the rationale and representative works of two most popular tracking frameworks in past ten years, i.e., the corelation filter and Siamese network for object tracking. Then we present some deep learning based tracking methods categorized by different network structures. We also introduce some classical strategies for handling the challenges in tracking problem. Further, this paper detailedly present and compare the benchmarks and challenges for tracking, from which we summarize the development history and development trend of visual tracking. Focusing on the future development of object tracking, which we think would be applied in real-world scenes before some problems to be addressed, such as the problems in long-term tracking, low-power high-speed tracking and attack-robust tracking. In the future, the integration of multimodal data, e.g., the depth image, thermal image with traditional color image, will provide more solutions for visual tracking. Moreover, tracking task will go together with some other tasks, e.g., video object detection and segmentation.

Likai Wang, Ruize Han, Wei Feng et al.

Gait recognition is an important AI task, which has been progressed rapidly with the development of deep learning. However, existing learning based gait recognition methods mainly focus on the single domain, especially the constrained laboratory environment. In this paper, we study a new problem of unsupervised domain adaptive gait recognition (UDA-GR), that learns a gait identifier with supervised labels from the indoor scenes (source domain), and is applied to the outdoor wild scenes (target domain). For this purpose, we develop an uncertainty estimation and regularization based UDA-GR method. Specifically, we investigate the characteristic of gaits in the indoor and outdoor scenes, for estimating the gait sample uncertainty, which is used in the unsupervised fine-tuning on the target domain to alleviate the noises of the pseudo labels. We also establish a new benchmark for the proposed problem, experimental results on which show the effectiveness of the proposed method. We will release the benchmark and source code in this work to the public.

Yupeng Zhang, Ruize Han, Ningnan Guo et al.

In real-world scenarios, continual changes in weather, illumination, and imaging conditions cause significant domain shifts, leading detectors trained on a single source domain to degrade severely in unseen environments. Existing single-domain generalized object detection (SDGOD) methods mainly rely on data augmentation or domain-invariant representation learning, but pay limited attention to detector mechanisms, leaving clear limitations under complex domain shifts. Through analytical experiments, we find that performance degradation is dominated by increasing missed detections, which fundamentally arises from reduced cross-domain stability of the detector: object-background and inter-instance relations become less stable in the encoding stage, while semantic-spatial alignment of query representations also becomes harder to maintain in the decoding stage. To this end, we propose VFM$^{4}$SDG, a dual-prior learning framework for SDGOD, which introduces a frozen vision foundation model (VFM) as a transferable cross-domain stability prior into detector representation learning and query modeling. In the encoding stage, we propose Cross-domain Stable Relational Prior Distillation to enhance the robustness of object-background and inter-instance relational modeling. In the decoding stage, we propose Semantic-Contextual Prior-based Query Enhancement, which injects category-level semantic prototypes and global visual context into queries to improve their semantic recognition and spatial localization stability in unseen domains. Extensive experiments show that the proposed method consistently outperforms existing SOTA methods on standard SDGOD benchmarks and two mainstream DETR-based detectors, demonstrating its effectiveness, robustness, and generality.

Likai Wang, Xiangqun Zhang, Ruize Han et al.

Person re-identification (Re-ID) is a classical computer vision task and has achieved great progress so far. Recently, long-term Re-ID with clothes-changing has attracted increasing attention. However, existing methods mainly focus on image-based setting, where richer temporal information is overlooked. In this paper, we focus on the relatively new yet practical problem of clothes-changing video-based person re-identification (CCVReID), which is less studied. We systematically study this problem by simultaneously considering the challenge of the clothes inconsistency issue and the temporal information contained in the video sequence for the person Re-ID problem. Based on this, we develop a two-branch confidence-aware re-ranking framework for handling the CCVReID problem. The proposed framework integrates two branches that consider both the classical appearance features and cloth-free gait features through a confidence-guided re-ranking strategy. This method provides the baseline method for further studies. Also, we build two new benchmark datasets for CCVReID problem, including a large-scale synthetic video dataset and a real-world one, both containing human sequences with various clothing changes. We will release the benchmark and code in this work to the public.

Lin Wang, Xiufen Ye, Donghao Zhang et al.

Three-dimensional (3D) images, such as CT, MRI, and PET, are common in medical imaging applications and important in clinical diagnosis. Semantic ambiguity is a typical feature of many medical image labels. It can be caused by many factors, such as the imaging properties, pathological anatomy, and the weak representation of the binary masks, which brings challenges to accurate 3D segmentation. In 2D medical images, using soft masks instead of binary masks generated by image matting to characterize lesions can provide rich semantic information, describe the structural characteristics of lesions more comprehensively, and thus benefit the subsequent diagnoses and analyses. In this work, we introduce image matting into the 3D scenes to describe the lesions in 3D medical images. The study of image matting in 3D modality is limited, and there is no high-quality annotated dataset related to 3D matting, therefore slowing down the development of data-driven deep-learning-based methods. To address this issue, we constructed the first 3D medical matting dataset and convincingly verified the validity of the dataset through quality control and downstream experiments in lung nodules classification. We then adapt the four selected state-of-the-art 2D image matting algorithms to 3D scenes and further customize the methods for CT images. Also, we propose the first end-to-end deep 3D matting network and implement a solid 3D medical image matting benchmark, which will be released to encourage further research.

Bo Zhang, Zuheng Ming, Wei Feng et al.

To benefit the complementary information between heterogeneous data, we introduce a new Multimodal Transformer (MMFormer) for Remote Sensing (RS) image classification using Hyperspectral Image (HSI) accompanied by another source of data such as Light Detection and Ranging (LiDAR). Compared with traditional Vision Transformer (ViT) lacking inductive biases of convolutions, we first introduce convolutional layers to our MMFormer to tokenize patches from multimodal data of HSI and LiDAR. Then we propose a Multi-scale Multi-head Self-Attention (MSMHSA) module to address the problem of compatibility which often limits to fuse HSI with high spectral resolution and LiDAR with relatively low spatial resolution. The proposed MSMHSA module can incorporate HSI to LiDAR data in a coarse-to-fine manner enabling us to learn a fine-grained representation. Extensive experiments on widely used benchmarks (e.g., Trento and MUUFL) demonstrate the effectiveness and superiority of our proposed MMFormer for RS image classification.

Lianyu Hu, Shenglan Liu, Wei Feng

It's common for current methods in skeleton-based action recognition to mainly consider capturing long-term temporal dependencies as skeleton sequences are typically long (>128 frames), which forms a challenging problem for previous approaches. In such conditions, short-term dependencies are few formally considered, which are critical for classifying similar actions. Most current approaches are consisted of interleaving spatial-only modules and temporal-only modules, where direct information flow among joints in adjacent frames are hindered, thus inferior to capture short-term motion and distinguish similar action pairs. To handle this limitation, we propose a general framework, coined as STGAT, to model cross-spacetime information flow. It equips the spatial-only modules with spatial-temporal modeling for regional perception. While STGAT is theoretically effective for spatial-temporal modeling, we propose three simple modules to reduce local spatial-temporal feature redundancy and further release the potential of STGAT, which (1) narrow the scope of self-attention mechanism, (2) dynamically weight joints along temporal dimension, and (3) separate subtle motion from static features, respectively. As a robust feature extractor, STGAT generalizes better upon classifying similar actions than previous methods, witnessed by both qualitative and quantitative results. STGAT achieves state-of-the-art performance on three large-scale datasets: NTU RGB+D 60, NTU RGB+D 120, and Kinetics Skeleton 400. Code is released.

Zekun Qian, Ruize Han, Wei Feng

Object-level spatial-temporal understanding is essential for video question answering, yet existing multimodal large language models (MLLMs) encode frames holistically and lack explicit mechanisms for fine-grained object grounding. Recent work addresses this by serializing bounding box coordinates as text tokens, but this text-coordinate paradigm suffers from a fundamental modality mismatch: object information is inherently visual, yet encoding it as text incurs a high token cost that forces aggressive temporal downsampling. We propose BoxTuning, which resolves this mismatch by injecting object spatial-temporal information directly into the visual modality. Colored bounding boxes and trajectory trails are rendered onto video frames as visual prompts, with only a concise color-to-object legend retained as text. This reduces the token cost significantly, achieving 87-93% text token reduction in practice. It also preserves full temporal resolution, where the trajectory trails further encode inter-frame motion direction and speed within each keyframe, recovering fine-grained dynamics that text-coordinate methods are forced to discard. Experimental results on five video QA benchmarks (CLEVRER, Perception Test, STAR, NExT-QA, IntentQA) show that BoxTuning surpasses text-coordinate baselines on spatially oriented tasks and nearly eliminates the accuracy degradation observed on reasoning-centric tasks, establishing visual prompting as a more natural and efficient paradigm for conveying object information to video MLLMs.

Cunling Bian, Wei Feng, Fanbo Meng et al.

Skeleton-based human action recognition has been drawing more interest recently due to its low sensitivity to appearance changes and the accessibility of more skeleton data. However, even the 3D skeletons captured in practice are still sensitive to the viewpoint and direction gave the occlusion of different human-body joints and the errors in human joint localization. Such view variance of skeleton data may significantly affect the performance of action recognition. To address this issue, we propose in this paper a new view-invariant representation learning approach, without any manual action labeling, for skeleton-based human action recognition. Specifically, we leverage the multi-view skeleton data simultaneously taken for the same person in the network training, by maximizing the mutual information between the representations extracted from different views, and then propose a global-local contrastive loss to model the multi-scale co-occurrence relationships in both spatial and temporal domains. Extensive experimental results show that the proposed method is robust to the view difference of the input skeleton data and significantly boosts the performance of unsupervised skeleton-based human action methods, resulting in new state-of-the-art accuracies on two challenging multi-view benchmarks of PKUMMD and NTU RGB+D.

Yu-Wei Zhan, Xin Wang, Hong Chen et al.

Video Large Language Models (Video LLMs) have shown impressive performance across a wide range of video-language tasks. However, they often fail in scenarios requiring a deeper understanding of physical dynamics. This limitation primarily arises from their reliance on appearance-based matching. Incorporating physical motion modeling is crucial for deeper video understanding, but presents three key challenges: (1) motion signals are often entangled with appearance variations, making it difficult to extract clean physical cues; (2) effective motion modeling requires not only continuous-time motion representations but also capturing physical dynamics; and (3) collecting accurate annotations for physical attributes is costly and often impractical. To address these issues, we propose PhyVLLM, a physical-guided video-language framework that explicitly incorporates physical motion into Video LLMs. Specifically, PhyVLLM disentangles visual appearance and object motion through a dual-branch encoder. To model physical dynamics over time, we incorporate a Neural Ordinary Differential Equation (Neural ODE) module, which generates differentiable physical dynamic representations. The resulting motion-aware representations are projected into the token space of a pretrained LLM, enabling physics reasoning without compromising the model's original multimodal capabilities. To circumvent the need for explicit physical labels, PhyVLLM employs a self-supervised manner to model the continuous evolution of object motion. Experimental results demonstrate that PhyVLLM significantly outperforms state-of-the-art Video LLMs on both physical reasoning and general video understanding tasks, highlighting the advantages of incorporating explicit physical modeling.

Zhenbang Du, Wei Feng, Haohan Wang et al.

In the e-commerce realm, compelling advertising images are pivotal for attracting customer attention. While generative models automate image generation, they often produce substandard images that may mislead customers and require significant labor costs to inspect. This paper delves into increasing the rate of available generated images. We first introduce a multi-modal Reliable Feedback Network (RFNet) to automatically inspect the generated images. Combining the RFNet into a recurrent process, Recurrent Generation, results in a higher number of available advertising images. To further enhance production efficiency, we fine-tune diffusion models with an innovative Consistent Condition regularization utilizing the feedback from RFNet (RFFT). This results in a remarkable increase in the available rate of generated images, reducing the number of attempts in Recurrent Generation, and providing a highly efficient production process without sacrificing visual appeal. We also construct a Reliable Feedback 1 Million (RF1M) dataset which comprises over one million generated advertising images annotated by human, which helps to train RFNet to accurately assess the availability of generated images and faithfully reflect the human feedback. Generally speaking, our approach offers a reliable solution for advertising image generation.

Wei Feng, Xin Wang, Hong Chen et al.

Video generation has witnessed great success recently, but their application in generating long videos still remains challenging due to the difficulty in maintaining the temporal consistency of generated videos and the high memory cost during generation. To tackle the problems, in this paper, we propose a brave and new idea of Multi-sentence Video Grounding for Long Video Generation, connecting the massive video moment retrieval to the video generation task for the first time, providing a new paradigm for long video generation. The method of our work can be summarized as three steps: (i) We design sequential scene text prompts as the queries for video grounding, utilizing the massive video moment retrieval to search for video moment segments that meet the text requirements in the video database. (ii) Based on the source frames of retrieved video moment segments, we adopt video editing methods to create new video content while preserving the temporal consistency of the retrieved video. Since the editing can be conducted segment by segment, and even frame by frame, it largely reduces the memory cost. (iii) We also attempt video morphing and personalized generation methods to improve the subject consistency of long video generation, providing ablation experimental results for the subtasks of long video generation. Our approach seamlessly extends the development in image/video editing, video morphing and personalized generation, and video grounding to the long video generation, offering effective solutions for generating long videos at low memory cost.

Tongkai Shi, Lianyu Hu, Fanhua Shang et al.

Sign language videos are an important medium for spreading and learning sign language. However, most existing human image synthesis methods produce sign language images with details that are distorted, blurred, or structurally incorrect. They also produce sign language video frames with poor temporal consistency, with anomalies such as flickering and abrupt detail changes between the previous and next frames. To address these limitations, we propose a novel Pose-Guided Motion Model (PGMM) for generating fine-grained and motion-consistent sign language videos. Firstly, we propose a new Coarse Motion Module (CMM), which completes the deformation of features by optical flow warping, thus transfering the motion of coarse-grained structures without changing the appearance; Secondly, we propose a new Pose Fusion Module (PFM), which guides the modal fusion of RGB and pose features, thus completing the fine-grained generation. Finally, we design a new metric, Temporal Consistency Difference (TCD) to quantitatively assess the degree of temporal consistency of a video by comparing the difference between the frames of the reconstructed video and the previous and next frames of the target video. Extensive qualitative and quantitative experiments show that our method outperforms state-of-the-art methods in most benchmark tests, with visible improvements in details and temporal consistency.

Run Ling, Ke Cao, Jian Lu et al.

Multi-subject video generation aims to synthesize videos from textual prompts and multiple reference images, ensuring that each subject preserves natural scale and visual fidelity. However, current methods face two challenges: scale inconsistency, where variations in subject size lead to unnatural generation, and permutation sensitivity, where the order of reference inputs causes subject distortion. In this paper, we propose MoFu, a unified framework that tackles both challenges. For scale inconsistency, we introduce Scale-Aware Modulation (SMO), an LLM-guided module that extracts implicit scale cues from the prompt and modulates features to ensure consistent subject sizes. To address permutation sensitivity, we present a simple yet effective Fourier Fusion strategy that processes the frequency information of reference features via the Fast Fourier Transform to produce a unified representation. Besides, we design a Scale-Permutation Stability Loss to jointly encourage scale-consistent and permutation-invariant generation. To further evaluate these challenges, we establish a dedicated benchmark with controlled variations in subject scale and reference permutation. Extensive experiments demonstrate that MoFu significantly outperforms existing methods in preserving natural scale, subject fidelity, and overall visual quality.

Tingliang Feng, Hao Shi, Xueyang Liu et al.

Many methods of semantic image segmentation have borrowed the success of open compound domain adaptation. They minimize the style gap between the images of source and target domains, more easily predicting the accurate pseudo annotations for target domain's images that train segmentation network. The existing methods globally adapt the scene style of the images, whereas the object styles of different categories or instances are adapted improperly. This paper proposes the Object Style Compensation, where we construct the Object-Level Discrepancy Memory with multiple sets of discrepancy features. The discrepancy features in a set capture the style changes of the same category's object instances adapted from target to source domains. We learn the discrepancy features from the images of source and target domains, storing the discrepancy features in memory. With this memory, we select appropriate discrepancy features for compensating the style information of the object instances of various categories, adapting the object styles to a unified style of source domain. Our method enables a more accurate computation of the pseudo annotations for target domain's images, thus yielding state-of-the-art results on different datasets.

Yueshan Lin, Wei Feng, Yunfei Chen et al.

This paper investigates latency-constrained resource synergization for mission-oriented non-terrestrial networks (NTNs) in post-disaster emergency scenarios. When terrestrial infrastructures are damaged, unmanned aerial vehicles (UAVs) equipped with edge information hubs (EIHs) are deployed to provide temporary coverage and synergize communication and computing resources for rapid situation awareness. We formulate a joint resource configuration and location optimization problem to minimize overall resource costs while guaranteeing stringent latency requirements. Through analytical derivations, we obtain closed-form optimal solutions that reveal the fundamental tradeoff between communication and computing resources, and develop a successive convex approximation method for EIH location optimization. Simulation results demonstrate that the proposed scheme achieves approximately 20% cost reduction compared with benchmark approaches, validating its optimality and effectiveness for mission-critical emergency response applications in the sixth-generation (6G) era.

Zhongxing Xu, Zhonghua Wang, Zhe Qian et al.

Recent advancements in multimodal large reasoning models (MLRMs) have significantly improved performance in visual question answering. However, we observe that transition words (e.g., because, however, and wait) are closely associated with hallucinations and tend to exhibit high-entropy states. We argue that adequate contextual reasoning information can be directly extracted from the token probability distribution. Inspired by superposed representation theory, we propose leveraging latent superposed reasoning to integrate multiple candidate semantics and maintain latent reasoning trajectories. The hypothesis is that reliance on discrete textual inputs may drive the model toward sequential explicit reasoning, underutilizing dense contextual cues during high-entropy reasoning stages. Therefore, we propose constructing rich semantic representations from the token probability distributions to enhance in-context reasoning. With this goal, we present Latent Entropy-Aware Decoding (LEAD), an efficient plug-and-play decoding strategy that leverages semantic context to achieve reliable reasoning. The heart of our method lies in entropy-aware reasoning mode switching. The model employs probability-weighted continuous embeddings under high-entropy states and transitions back to discrete token embeddings as entropy decreases. Moreover, we propose a prior-guided visual anchor injection strategy that encourages the model to focus on visual information. Extensive experiments show that LEAD effectively mitigates hallucinations across various MLRMs on multiple benchmarks.

Zhicong Lu, Zichuan Lin, Wei Jia et al.

While large language models excel in diverse domains, their performance on complex longhorizon agentic decision-making tasks remains limited. Most existing methods concentrate on designing effective reward models (RMs) to advance performance via multi-turn reinforcement learning. However, they suffer from delayed propagation in sparse outcome rewards and unreliable credit assignment with potentially overly fine-grained and unfocused turnlevel process rewards. In this paper, we propose (HISR) exploiting Hindsight Information to modulate Segmental process Rewards, which closely aligns rewards with sub-goals and underscores significant segments to enhance the reliability of credit assignment. Specifically, a segment-level process RM is presented to assign rewards for each sub-goal in the task, avoiding excessively granular allocation to turns. To emphasize significant segments in the trajectory, a hindsight model is devised to reflect the preference of performing a certain action after knowing the trajectory outcome. With this characteristic, we design the ratios of sequence likelihoods between hindsight and policy model to measure action importance. The ratios are subsequently employed to aggregate segment importance scores, which in turn modulate segmental process rewards, enhancing credit assignment reliability. Extensive experimental results on three publicly benchmarks demonstrate the validity of our method.

Feilong Tang, Zhongxing Xu, Zhaojun Qu et al.

Recent weakly supervised semantic segmentation (WSSS) methods strive to incorporate contextual knowledge to improve the completeness of class activation maps (CAM). In this work, we argue that the knowledge bias between instances and contexts affects the capability of the prototype to sufficiently understand instance semantics. Inspired by prototype learning theory, we propose leveraging prototype awareness to capture diverse and fine-grained feature attributes of instances. The hypothesis is that contextual prototypes might erroneously activate similar and frequently co-occurring object categories due to this knowledge bias. Therefore, we propose to enhance the prototype representation ability by mitigating the bias to better capture spatial coverage in semantic object regions. With this goal, we present a Context Prototype-Aware Learning (CPAL) strategy, which leverages semantic context to enrich instance comprehension. The core of this method is to accurately capture intra-class variations in object features through context-aware prototypes, facilitating the adaptation to the semantic attributes of various instances. We design feature distribution alignment to optimize prototype awareness, aligning instance feature distributions with dense features. In addition, a unified training framework is proposed to combine label-guided classification supervision and prototypes-guided self-supervision. Experimental results on PASCAL VOC 2012 and MS COCO 2014 show that CPAL significantly improves off-the-shelf methods and achieves state-of-the-art performance. The project is available at https://github.com/Barrett-python/CPAL.

Zekun Qian, Ruize Han, Wei Feng et al.

We study a novel yet practical problem of open-corpus multi-object tracking (OCMOT), which extends the MOT into localizing, associating, and recognizing generic-category objects of both seen (base) and unseen (novel) classes, but without the category text list as prompt. To study this problem, the top priority is to build a benchmark. In this work, we build OCTrackB, a large-scale and comprehensive benchmark, to provide a standard evaluation platform for the OCMOT problem. Compared to previous datasets, OCTrackB has more abundant and balanced base/novel classes and the corresponding samples for evaluation with less bias. We also propose a new multi-granularity recognition metric to better evaluate the generative object recognition in OCMOT. By conducting the extensive benchmark evaluation, we report and analyze the results of various state-of-the-art methods, which demonstrate the rationale of OCMOT, as well as the usefulness and advantages of OCTrackB.

Weiqi Li, Fan Lyu, Fanhua Shang et al.

Real-world data is extremely imbalanced and presents a long-tailed distribution, resulting in models that are biased towards classes with sufficient samples and perform poorly on rare classes. Recent methods propose to rebalance classes but they undertake the seesaw dilemma (what is increasing performance on tail classes may decrease that of head classes, and vice versa). In this paper, we argue that the seesaw dilemma is derived from gradient imbalance of different classes, in which gradients of inappropriate classes are set to important for updating, thus are prone to overcompensation or undercompensation on tail classes. To achieve ideal compensation, we formulate the long-tailed recognition as an multi-objective optimization problem, which fairly respects the contributions of head and tail classes simultaneously. For efficiency, we propose a Gradient-Balancing Grouping (GBG) strategy to gather the classes with similar gradient directions, thus approximately make every update under a Pareto descent direction. Our GBG method drives classes with similar gradient directions to form more representative gradient and provide ideal compensation to the tail classes. Moreover, We conduct extensive experiments on commonly used benchmarks in long-tailed learning and demonstrate the superiority of our method over existing SOTA methods.

Zekun Qian, Wei Feng, Ruize Han et al.

Multi-Object Tracking (MOT) has traditionally focused on a few specific categories, restricting its applicability to real-world scenarios involving diverse objects. Open-Vocabulary Multi-Object Tracking (OVMOT) addresses this by enabling tracking of arbitrary categories, including novel objects unseen during training. However, current progress is constrained by two challenges: the lack of continuously annotated video data for training, and the lack of a customized OVMOT framework to synergistically handle detection and association. We address the data bottleneck by constructing C-TAO, the first continuously annotated training set for OVMOT, which increases annotation density by 26x over the original TAO and captures smooth motion dynamics and intermediate object states. For the framework bottleneck, we propose COVTrack++, a synergistic framework that achieves a bidirectional reciprocal mechanism between detection and association through three modules: (1) Multi-Cue Adaptive Fusion (MCF) dynamically balances appearance, motion, and semantic cues for association feature learning; (2) Multi-Granularity Hierarchical Aggregation (MGA) exploits hierarchical spatial relationships in dense detections, where visible child nodes (e.g., object parts) assist occluded parent objects (e.g., whole body) for association feature enhancement; (3) Temporal Confidence Propagation (TCP) recovers flickering detections through high-confidence tracked objects boosting low-confidence candidates across frames, stabilizing trajectories. Extensive experiments on TAO demonstrate state-of-the-art performance, with novel TETA reaching 35.4% and 30.5% on validation and test sets, improving novel AssocA by 4.8% and novel LocA by 5.8% over previous methods, and show strong zero-shot generalization on BDD100K. The code and dataset will be publicly available.