Jiangning Zhang

Jianzong Wu, Xiangtai Li, Shilin Xu et al.

In the field of visual scene understanding, deep neural networks have made impressive advancements in various core tasks like segmentation, tracking, and detection. However, most approaches operate on the close-set assumption, meaning that the model can only identify pre-defined categories that are present in the training set. Recently, open vocabulary settings were proposed due to the rapid progress of vision language pre-training. These new approaches seek to locate and recognize categories beyond the annotated label space. The open vocabulary approach is more general, practical, and effective compared to weakly supervised and zero-shot settings. This paper provides a thorough review of open vocabulary learning, summarizing and analyzing recent developments in the field. In particular, we begin by comparing it to related concepts such as zero-shot learning, open-set recognition, and out-of-distribution detection. Then, we review several closely related tasks in the case of segmentation and detection, including long-tail problems, few-shot, and zero-shot settings. For the method survey, we first present the basic knowledge of detection and segmentation in close-set as the preliminary knowledge. Next, we examine various scenarios in which open vocabulary learning is used, identifying common design elements and core ideas. Then, we compare the recent detection and segmentation approaches in commonly used datasets and benchmarks. Finally, we conclude with insights, issues, and discussions regarding future research directions. To our knowledge, this is the first comprehensive literature review of open vocabulary learning. We keep tracing related works at https://github.com/jianzongwu/Awesome-Open-Vocabulary.

Yufei Liang, Jiangning Zhang, Shiwei Zhao et al.

Density-based and classification-based methods have ruled unsupervised anomaly detection in recent years, while reconstruction-based methods are rarely mentioned for the poor reconstruction ability and low performance. However, the latter requires no costly extra training samples for the unsupervised training that is more practical, so this paper focuses on improving this kind of method and proposes a novel Omni-frequency Channel-selection Reconstruction (OCR-GAN) network to handle anomaly detection task in a perspective of frequency. Concretely, we propose a Frequency Decoupling (FD) module to decouple the input image into different frequency components and model the reconstruction process as a combination of parallel omni-frequency image restorations, as we observe a significant difference in the frequency distribution of normal and abnormal images. Given the correlation among multiple frequencies, we further propose a Channel Selection (CS) module that performs frequency interaction among different encoders by adaptively selecting different channels. Abundant experiments demonstrate the effectiveness and superiority of our approach over different kinds of methods, e.g., achieving a new state-of-the-art 98.3 detection AUC on the MVTec AD dataset without extra training data that markedly surpasses the reconstruction-based baseline by +38.1 and the current SOTA method by +0.3. Source code is available at https://github.com/zhangzjn/OCR-GAN.

Teng Hu, Jiangning Zhang, Liang Liu et al. · tsinghua

Training a generative model with limited number of samples is a challenging task. Current methods primarily rely on few-shot model adaption to train the network. However, in scenarios where data is extremely limited (less than 10), the generative network tends to overfit and suffers from content degradation. To address these problems, we propose a novel phasic content fusing few-shot diffusion model with directional distribution consistency loss, which targets different learning objectives at distinct training stages of the diffusion model. Specifically, we design a phasic training strategy with phasic content fusion to help our model learn content and style information when t is large, and learn local details of target domain when t is small, leading to an improvement in the capture of content, style and local details. Furthermore, we introduce a novel directional distribution consistency loss that ensures the consistency between the generated and source distributions more efficiently and stably than the prior methods, preventing our model from overfitting. Finally, we propose a cross-domain structure guidance strategy that enhances structure consistency during domain adaptation. Theoretical analysis, qualitative and quantitative experiments demonstrate the superiority of our approach in few-shot generative model adaption tasks compared to state-of-the-art methods. The source code is available at: https://github.com/sjtuplayer/few-shot-diffusion.

Yue Wang, Jinlong Peng, Jiangning Zhang et al. · tsinghua

Face analysis tasks have a wide range of applications, but the universal facial representation has only been explored in a few works. In this paper, we explore high-performance pre-training methods to boost the face analysis tasks such as face alignment and face parsing. We propose a self-supervised pre-training framework, called \textbf{\it Mask Contrastive Face (MCF)}, with mask image modeling and a contrastive strategy specially adjusted for face domain tasks. To improve the facial representation quality, we use feature map of a pre-trained visual backbone as a supervision item and use a partially pre-trained decoder for mask image modeling. To handle the face identity during the pre-training stage, we further use random masks to build contrastive learning pairs. We conduct the pre-training on the LAION-FACE-cropped dataset, a variants of LAION-FACE 20M, which contains more than 20 million face images from Internet websites. For efficiency pre-training, we explore our framework pre-training performance on a small part of LAION-FACE-cropped and verify the superiority with different pre-training settings. Our model pre-trained with the full pre-training dataset outperforms the state-of-the-art methods on multiple downstream tasks. Our model achieves 0.932 NME$_{diag}$ for AFLW-19 face alignment and 93.96 F1 score for LaPa face parsing. Code is available at https://github.com/nomewang/MCF.

Yue Wang, Jinlong Peng, Jiangning Zhang et al.

2D-based Industrial Anomaly Detection has been widely discussed, however, multimodal industrial anomaly detection based on 3D point clouds and RGB images still has many untouched fields. Existing multimodal industrial anomaly detection methods directly concatenate the multimodal features, which leads to a strong disturbance between features and harms the detection performance. In this paper, we propose Multi-3D-Memory (M3DM), a novel multimodal anomaly detection method with hybrid fusion scheme: firstly, we design an unsupervised feature fusion with patch-wise contrastive learning to encourage the interaction of different modal features; secondly, we use a decision layer fusion with multiple memory banks to avoid loss of information and additional novelty classifiers to make the final decision. We further propose a point feature alignment operation to better align the point cloud and RGB features. Extensive experiments show that our multimodal industrial anomaly detection model outperforms the state-of-the-art (SOTA) methods on both detection and segmentation precision on MVTec-3D AD dataset. Code is available at https://github.com/nomewang/M3DM.

Jiangning Zhang, Xiangtai Li, Yabiao Wang et al.

Motivated by biological evolution, this paper explains the rationality of Vision Transformer by analogy with the proven practical evolutionary algorithm (EA) and derives that both have consistent mathematical formulation. Then inspired by effective EA variants, we propose a novel pyramid EATFormer backbone that only contains the proposed EA-based transformer (EAT) block, which consists of three residual parts, i.e., Multi-scale region aggregation, global and local interaction, and feed-forward network modules, to model multi-scale, interactive, and individual information separately. Moreover, we design a task-related head docked with transformer backbone to complete final information fusion more flexibly and improve a modulated deformable MSA to dynamically model irregular locations. Massive quantitative and quantitative experiments on image classification, downstream tasks, and explanatory experiments demonstrate the effectiveness and superiority of our approach over state-of-the-art methods. E.g., our Mobile (1.8 M), Tiny (6.1 M), Small (24.3 M), and Base (49.0 M) models achieve 69.4, 78.4, 83.1, and 83.9 Top-1 only trained on ImageNet-1K with naive training recipe; EATFormer-Tiny/Small/Base armed Mask-R-CNN obtain 45.4/47.4/49.0 box AP and 41.4/42.9/44.2 mask AP on COCO detection, surpassing contemporary MPViT-T, Swin-T, and Swin-S by 0.6/1.4/0.5 box AP and 0.4/1.3/0.9 mask AP separately with less FLOPs; Our EATFormer-Small/Base achieve 47.3/49.3 mIoU on ADE20K by Upernet that exceeds Swin-T/S by 2.8/1.7. Code is available at https://github.com/zhangzjn/EATFormer.

Xiangtai Li, Jiangning Zhang, Yibo Yang et al.

In this paper, we focus on exploring effective methods for faster and accurate semantic segmentation. A common practice to improve the performance is to attain high-resolution feature maps with strong semantic representation. Two strategies are widely used: atrous convolutions and feature pyramid fusion, while both are either computationally intensive or ineffective. Inspired by the Optical Flow for motion alignment between adjacent video frames, we propose a Flow Alignment Module (FAM) to learn \textit{Semantic Flow} between feature maps of adjacent levels and broadcast high-level features to high-resolution features effectively and efficiently. Furthermore, integrating our FAM to a standard feature pyramid structure exhibits superior performance over other real-time methods, even on lightweight backbone networks, such as ResNet-18 and DFNet. Then to further speed up the inference procedure, we also present a novel Gated Dual Flow Alignment Module to directly align high-resolution feature maps and low-resolution feature maps where we term the improved version network as SFNet-Lite. Extensive experiments are conducted on several challenging datasets, where results show the effectiveness of both SFNet and SFNet-Lite. In particular, when using Cityscapes test set, the SFNet-Lite series achieve 80.1 mIoU while running at 60 FPS using ResNet-18 backbone and 78.8 mIoU while running at 120 FPS using STDC backbone on RTX-3090. Moreover, we unify four challenging driving datasets into one large dataset, which we named Unified Driving Segmentation (UDS) dataset. It contains diverse domain and style information. We benchmark several representative works on UDS. Both SFNet and SFNet-Lite still achieve the best speed and accuracy trade-off on UDS, which serves as a strong baseline in such a challenging setting. The code and models are publicly available at https://github.com/lxtGH/SFSegNets.

Liang Liu, Boshen Zhang, Jiangning Zhang et al.

Scale variation across object instances remains a key challenge in object detection task. Despite the remarkable progress made by modern detection models, this challenge is particularly evident in the semi-supervised case. While existing semi-supervised object detection methods rely on strict conditions to filter high-quality pseudo labels from network predictions, we observe that objects with extreme scale tend to have low confidence, resulting in a lack of positive supervision for these objects. In this paper, we propose a novel framework that addresses the scale variation problem by introducing a mixed scale teacher to improve pseudo label generation and scale-invariant learning. Additionally, we propose mining pseudo labels using score promotion of predictions across scales, which benefits from better predictions from mixed scale features. Our extensive experiments on MS COCO and PASCAL VOC benchmarks under various semi-supervised settings demonstrate that our method achieves new state-of-the-art performance. The code and models are available at \url{https://github.com/lliuz/MixTeacher}.

Jiangning Zhang, Haoyang He, Xuhai Chen et al.

Large Multimodal Model (LMM) GPT-4V(ision) endows GPT-4 with visual grounding capabilities, making it possible to handle certain tasks through the Visual Question Answering (VQA) paradigm. This paper explores the potential of VQA-oriented GPT-4V in the recently popular visual Anomaly Detection (AD) and is the first to conduct qualitative and quantitative evaluations on the popular MVTec AD and VisA datasets. Considering that this task requires both image-/pixel-level evaluations, the proposed GPT-4V-AD framework contains three components: \textbf{\textit{1)}} Granular Region Division, \textbf{\textit{2)}} Prompt Designing, \textbf{\textit{3)}} Text2Segmentation for easy quantitative evaluation, and have made some different attempts for comparative analysis. The results show that GPT-4V can achieve certain results in the zero-shot AD task through a VQA paradigm, such as achieving image-level 77.1/88.0 and pixel-level 68.0/76.6 AU-ROCs on MVTec AD and VisA datasets, respectively. However, its performance still has a certain gap compared to the state-of-the-art zero-shot method, \eg, WinCLIP and CLIP-AD, and further researches are needed. This study provides a baseline reference for the research of VQA-oriented LMM in the zero-shot AD task, and we also post several possible future works. Code is available at \url{https://github.com/zhangzjn/GPT-4V-AD}.

Haohan Wang, Liang Liu, Wuhao Zhang et al.

Few-shot semantic segmentation aims to learn to segment unseen class objects with the guidance of only a few support images. Most previous methods rely on the pixel-level label of support images. In this paper, we focus on a more challenging setting, in which only the image-level labels are available. We propose a general framework to firstly generate coarse masks with the help of the powerful vision-language model CLIP, and then iteratively and mutually refine the mask predictions of support and query images. Extensive experiments on PASCAL-5i and COCO-20i datasets demonstrate that our method not only outperforms the state-of-the-art weakly supervised approaches by a significant margin, but also achieves comparable or better results to recent supervised methods. Moreover, our method owns an excellent generalization ability for the images in the wild and uncommon classes. Code will be available at https://github.com/Whileherham/IMR-HSNet.

Yue Han, Jiangning Zhang, Yabiao Wang et al.

Few-Shot Instance Segmentation (FSIS) requires detecting and segmenting novel classes with limited support examples. Existing methods based on Region Proposal Networks (RPNs) face two issues: 1) Overfitting suppresses novel class objects; 2) Dual-branch models require complex spatial correlation strategies to prevent spatial information loss when generating class prototypes. We introduce a unified framework, Reference Twice (RefT), to exploit the relationship between support and query features for FSIS and related tasks. Our three main contributions are: 1) A novel transformer-based baseline that avoids overfitting, offering a new direction for FSIS; 2) Demonstrating that support object queries encode key factors after base training, allowing query features to be enhanced twice at both feature and query levels using simple cross-attention, thus avoiding complex spatial correlation interaction; 3) Introducing a class-enhanced base knowledge distillation loss to address the issue of DETR-like models struggling with incremental settings due to the input projection layer, enabling easy extension to incremental FSIS. Extensive experimental evaluations on the COCO dataset under three FSIS settings demonstrate that our method performs favorably against existing approaches across different shots, \eg, $+8.2/+9.4$ performance gain over state-of-the-art methods with 10/30-shots. Source code and models will be available at https://github.com/hanyue1648/RefT.

Yunkang Cao, Jiangning Zhang, Luca Frittoli et al.

Zero-shot anomaly detection (ZSAD) targets the identification of anomalies within images from arbitrary novel categories. This study introduces AdaCLIP for the ZSAD task, leveraging a pre-trained vision-language model (VLM), CLIP. AdaCLIP incorporates learnable prompts into CLIP and optimizes them through training on auxiliary annotated anomaly detection data. Two types of learnable prompts are proposed: static and dynamic. Static prompts are shared across all images, serving to preliminarily adapt CLIP for ZSAD. In contrast, dynamic prompts are generated for each test image, providing CLIP with dynamic adaptation capabilities. The combination of static and dynamic prompts is referred to as hybrid prompts, and yields enhanced ZSAD performance. Extensive experiments conducted across 14 real-world anomaly detection datasets from industrial and medical domains indicate that AdaCLIP outperforms other ZSAD methods and can generalize better to different categories and even domains. Finally, our analysis highlights the importance of diverse auxiliary data and optimized prompts for enhanced generalization capacity. Code is available at https://github.com/caoyunkang/AdaCLIP.

Haohan Wang, Liang Liu, Boshen Zhang et al.

Fully supervised object detection requires training images in which all instances are annotated. This is actually impractical due to the high labor and time costs and the unavoidable missing annotations. As a result, the incomplete annotation in each image could provide misleading supervision and harm the training. Recent works on sparsely annotated object detection alleviate this problem by generating pseudo labels for the missing annotations. Such a mechanism is sensitive to the threshold of the pseudo label score. However, the effective threshold is different in different training stages and among different object detectors. Therefore, the current methods with fixed thresholds have sub-optimal performance, and are difficult to be applied to other detectors. In order to resolve this obstacle, we propose a Calibrated Teacher, of which the confidence estimation of the prediction is well calibrated to match its real precision. In this way, different detectors in different training stages would share a similar distribution of the output confidence, so that multiple detectors could share the same fixed threshold and achieve better performance. Furthermore, we present a simple but effective Focal IoU Weight (FIoU) for the classification loss. FIoU aims at reducing the loss weight of false negative samples caused by the missing annotation, and thus works as the complement of the teacher-student paradigm. Extensive experiments show that our methods set new state-of-the-art under all different sparse settings in COCO. Code will be available at https://github.com/Whileherham/CalibratedTeacher.

Yuanpeng Tu, Yuxi Li, Boshen Zhang et al.

Robust autonomous driving requires agents to accurately identify unexpected areas (anomalies) in urban scenes. To this end, some critical issues remain open: how to design advisable metric to measure anomalies, and how to properly generate training samples of anomaly data? Classical effort in anomaly detection usually resorts to pixel-wise uncertainty or sample synthesis, which ignores the contextual information and sometimes requires auxiliary data with fine-grained annotations. On the contrary, in this paper, we exploit the strong context-dependent nature of the segmentation task and design an energy-guided self-supervised framework for anomaly segmentation, which optimizes an anomaly head by maximizing the likelihood of self-generated anomaly pixels. For this purpose, we design two estimators to model anomaly likelihood, one is a task-agnostic binary estimator and the other depicts the likelihood as residual of task-oriented joint energy. Based on the proposed estimators, we devise an adaptive self-supervised training framework, which exploits the contextual reliance and estimated likelihood to refine mask annotations in anomaly areas. We conduct extensive experiments on challenging Fishyscapes and Road Anomaly benchmarks, demonstrating that without any auxiliary data or synthetic models, our method can still achieve comparable performance to supervised competitors. Code is available at https://github.com/yuanpengtu/SLEEG..

Jiangning Zhang, Xiangtai Li, Jian Li et al.

This paper focuses on developing modern, efficient, lightweight models for dense predictions while trading off parameters, FLOPs, and performance. Inverted Residual Block (IRB) serves as the infrastructure for lightweight CNNs, but no counterpart has been recognized by attention-based studies. This work rethinks lightweight infrastructure from efficient IRB and effective components of Transformer from a unified perspective, extending CNN-based IRB to attention-based models and abstracting a one-residual Meta Mobile Block (MMB) for lightweight model design. Following simple but effective design criterion, we deduce a modern Inverted Residual Mobile Block (iRMB) and build a ResNet-like Efficient MOdel (EMO) with only iRMB for down-stream tasks. Extensive experiments on ImageNet-1K, COCO2017, and ADE20K benchmarks demonstrate the superiority of our EMO over state-of-the-art methods, e.g., EMO-1M/2M/5M achieve 71.5, 75.1, and 78.4 Top-1 that surpass equal-order CNN-/Attention-based models, while trading-off the parameter, efficiency, and accuracy well: running 2.8-4.0x faster than EdgeNeXt on iPhone14.

Haoyang He, Jiangning Zhang, Guanzhong Tian et al.

This study explores the recently proposed and challenging multi-view Anomaly Detection (AD) task. Single-view tasks will encounter blind spots from other perspectives, resulting in inaccuracies in sample-level prediction. Therefore, we introduce the Multi-View Anomaly Detection (MVAD) approach, which learns and integrates features from multi-views. Specifically, we propose a Multi-View Adaptive Selection (MVAS) algorithm for feature learning and fusion across multiple views. The feature maps are divided into neighbourhood attention windows to calculate a semantic correlation matrix between single-view windows and all other views, which is an attention mechanism conducted for each single-view window and the top-k most correlated multi-view windows. Adjusting the window sizes and top-k can minimise the complexity to O((hw)^4/3). Extensive experiments on the Real-IAD dataset under the multi-class setting validate the effectiveness of our approach, achieving state-of-the-art performance with an average improvement of +2.5 across 10 metrics at the sample/image/pixel levels, using only 18M parameters and requiring fewer FLOPs and training time. The codes are available at https://github.com/lewandofskee/MVAD.

Xuhai Chen, Jiangning Zhang, Chao Xu et al.

Most of the existing blind image Super-Resolution (SR) methods assume that the blur kernels are space-invariant. However, the blur involved in real applications are usually space-variant due to object motion, out-of-focus, etc., resulting in severe performance drop of the advanced SR methods. To address this problem, we firstly introduce two new datasets with out-of-focus blur, i.e., NYUv2-BSR and Cityscapes-BSR, to support further researches of blind SR with space-variant blur. Based on the datasets, we design a novel Cross-MOdal fuSion network (CMOS) that estimate both blur and semantics simultaneously, which leads to improved SR results. It involves a feature Grouping Interactive Attention (GIA) module to make the two modalities interact more effectively and avoid inconsistency. GIA can also be used for the interaction of other features because of the universality of its structure. Qualitative and quantitative experiments compared with state-of-the-art methods on above datasets and real-world images demonstrate the superiority of our method, e.g., obtaining PSNR/SSIM by +1.91/+0.0048 on NYUv2-BSR than MANet.

Tong Zhang, Jiangning Zhang, Zhucun Xue et al.

Balancing convergence speed, generalization capability, and computational efficiency remains a core challenge in deep learning optimization. First-order gradient descent methods, epitomized by stochastic gradient descent (SGD) and Adam, serve as the cornerstone of modern training pipelines. However, large-scale model training, stringent differential privacy requirements, and distributed learning paradigms expose critical limitations in these conventional approaches regarding privacy protection and memory efficiency. To mitigate these bottlenecks, researchers explore second-order optimization techniques to surpass first-order performance ceilings, while zeroth-order methods reemerge to alleviate memory constraints inherent to large-scale training. Despite this proliferation of methodologies, the field lacks a cohesive framework that unifies underlying principles and delineates application scenarios for these disparate approaches. In this work, we retrospectively analyze the evolutionary trajectory of deep learning optimization algorithms and present a comprehensive empirical evaluation of mainstream optimizers across diverse model architectures and training scenarios. We distill key emerging trends and fundamental design trade-offs, pinpointing promising directions for future research. By synthesizing theoretical insights with extensive empirical evidence, we provide actionable guidance for designing next-generation highly efficient, robust, and trustworthy optimization methods. The code is available at https://github.com/APRIL-AIGC/Awesome-Optimizer.

Xuhai Chen, Jiangning Zhang, Guanzhong Tian et al.

This paper considers zero-shot Anomaly Detection (AD), performing AD without reference images of the test objects. We propose a framework called CLIP-AD to leverage the zero-shot capabilities of the large vision-language model CLIP. Firstly, we reinterpret the text prompts design from a distributional perspective and propose a Representative Vector Selection (RVS) paradigm to obtain improved text features. Secondly, we note opposite predictions and irrelevant highlights in the direct computation of the anomaly maps. To address these issues, we introduce a Staged Dual-Path model (SDP) that leverages features from various levels and applies architecture and feature surgery. Lastly, delving deeply into the two phenomena, we point out that the image and text features are not aligned in the joint embedding space. Thus, we introduce a fine-tuning strategy by adding linear layers and construct an extended model SDP+, further enhancing the performance. Abundant experiments demonstrate the effectiveness of our approach, e.g., on MVTec-AD, SDP outperforms the SOTA WinCLIP by +4.2/+10.7 in segmentation metrics F1-max/PRO, while SDP+ achieves +8.3/+20.5 improvements.

Chao Xu, Jiangning Zhang, Miao Hua et al.

This paper presents a novel Region-Aware Face Swapping (RAFSwap) network to achieve identity-consistent harmonious high-resolution face generation in a local-global manner: \textbf{1)} Local Facial Region-Aware (FRA) branch augments local identity-relevant features by introducing the Transformer to effectively model misaligned cross-scale semantic interaction. \textbf{2)} Global Source Feature-Adaptive (SFA) branch further complements global identity-relevant cues for generating identity-consistent swapped faces. Besides, we propose a \textit{Face Mask Predictor} (FMP) module incorporated with StyleGAN2 to predict identity-relevant soft facial masks in an unsupervised manner that is more practical for generating harmonious high-resolution faces. Abundant experiments qualitatively and quantitatively demonstrate the superiority of our method for generating more identity-consistent high-resolution swapped faces over SOTA methods, \eg, obtaining 96.70 ID retrieval that outperforms SOTA MegaFS by 5.87$\uparrow$.

Yuanpeng Tu, Boshen Zhang, Yuxi Li et al.

Collecting large-scale datasets is crucial for training deep models, annotating the data, however, inevitably yields noisy labels, which poses challenges to deep learning algorithms. Previous efforts tend to mitigate this problem via identifying and removing noisy samples or correcting their labels according to the statistical properties (e.g., loss values) among training samples. In this paper, we aim to tackle this problem from a new perspective, delving into the deep feature maps, we empirically find that models trained with clean and mislabeled samples manifest distinguishable activation feature distributions. From this observation, a novel robust training approach termed adversarial noisy masking is proposed. The idea is to regularize deep features with a label quality guided masking scheme, which adaptively modulates the input data and label simultaneously, preventing the model to overfit noisy samples. Further, an auxiliary task is designed to reconstruct input data, it naturally provides noise-free self-supervised signals to reinforce the generalization ability of deep models. The proposed method is simple and flexible, it is tested on both synthetic and real-world noisy datasets, where significant improvements are achieved over previous state-of-the-art methods.

Xiaofeng Mao, Zhengkai Jiang, Fu-Yun Wang et al.

Video diffusion models have shown great potential in generating high-quality videos, making them an increasingly popular focus. However, their inherent iterative nature leads to substantial computational and time costs. While efforts have been made to accelerate video diffusion by reducing inference steps (through techniques like consistency distillation) and GAN training (these approaches often fall short in either performance or training stability). In this work, we introduce a two-stage training framework that effectively combines consistency distillation with GAN training to address these challenges. Additionally, we propose a novel video discriminator design, which eliminates the need for decoding the video latents and improves the final performance. Our model is capable of producing high-quality videos in merely one-step, with the flexibility to perform multi-step refinement for further performance enhancement. Our quantitative evaluation on the OpenWebVid-1M benchmark shows that our model significantly outperforms existing methods. Notably, our 1-step performance(FVD 171.15) exceeds the 8-step performance of the consistency distillation based method, AnimateLCM (FVD 184.79), and approaches the 25-step performance of advanced Stable Video Diffusion (FVD 156.94).

Jiafu Wu, Jian Li, Jiangning Zhang et al.

Convolution-based and Transformer-based vision backbone networks process images into the grid or sequence structures, respectively, which are inflexible for capturing irregular objects. Though Vision GNN (ViG) adopts graph-level features for complex images, it has some issues, such as inaccurate neighbor node selection, expensive node information aggregation calculation, and over-smoothing in the deep layers. To address the above problems, we propose a Progressive Vision Graph (PVG) architecture for vision recognition task. Compared with previous works, PVG contains three main components: 1) Progressively Separated Graph Construction (PSGC) to introduce second-order similarity by gradually increasing the channel of the global graph branch and decreasing the channel of local branch as the layer deepens; 2) Neighbor nodes information aggregation and update module by using Max pooling and mathematical Expectation (MaxE) to aggregate rich neighbor information; 3) Graph error Linear Unit (GraphLU) to enhance low-value information in a relaxed form to reduce the compression of image detail information for alleviating the over-smoothing. Extensive experiments on mainstream benchmarks demonstrate the superiority of PVG over state-of-the-art methods, e.g., our PVG-S obtains 83.0% Top-1 accuracy on ImageNet-1K that surpasses GNN-based ViG-S by +0.9 with the parameters reduced by 18.5%, while the largest PVG-B obtains 84.2% that has +0.5 improvement than ViG-B. Furthermore, our PVG-S obtains +1.3 box AP and +0.4 mask AP gains than ViG-S on COCO dataset.

Jiangning Zhang, Junwei Zhu, Zhenye Gan et al.

We propose a multimodal-driven framework for high-fidelity long-term digital human animation termed $\textbf{Soul}$, which generates semantically coherent videos from a single-frame portrait image, text prompts, and audio, achieving precise lip synchronization, vivid facial expressions, and robust identity preservation. We construct Soul-1M, containing 1 million finely annotated samples with a precise automated annotation pipeline (covering portrait, upper-body, full-body, and multi-person scenes) to mitigate data scarcity, and we carefully curate Soul-Bench for comprehensive and fair evaluation of audio-/text-guided animation methods. The model is built on the Wan2.2-5B backbone, integrating audio-injection layers and multiple training strategies together with threshold-aware codebook replacement to ensure long-term generation consistency. Meanwhile, step/CFG distillation and a lightweight VAE are used to optimize inference efficiency, achieving an 11.4$\times$ speedup with negligible quality loss. Extensive experiments show that Soul significantly outperforms current leading open-source and commercial models on video quality, video-text alignment, identity preservation, and lip-synchronization accuracy, demonstrating broad applicability in real-world scenarios such as virtual anchors and film production. Project page at https://zhangzjn.github.io/projects/Soul/

Haozhen Gong, Xiaozhong Ji, Yuansen Liu et al.

MLLMs MLLMs are beginning to appear in clinical workflows, but their ability to perform complex medical reasoning remains unclear. We present Med-CMR, a fine-grained Medical Complex Multimodal Reasoning benchmark. Med-CMR distinguishes from existing counterparts by three core features: 1) Systematic capability decomposition, splitting medical multimodal reasoning into fine-grained visual understanding and multi-step reasoning to enable targeted evaluation; 2) Challenging task design, with visual understanding across three key dimensions (small-object detection, fine-detail discrimination, spatial understanding) and reasoning covering four clinically relevant scenarios (temporal prediction, causal reasoning, long-tail generalization, multi-source integration); 3) Broad, high-quality data coverage, comprising 20,653 Visual Question Answering (VQA) pairs spanning 11 organ systems and 12 imaging modalities, validated via a rigorous two-stage (human expert + model-assisted) review to ensure clinical authenticity. We evaluate 18 state-of-the-art MLLMs with Med-CMR, revealing GPT-5 as the top-performing commercial model: 57.81 accuracy on multiple-choice questions (MCQs) and a 48.70 open-ended score, outperforming Gemini 2.5 Pro (49.87 MCQ accuracy, 45.98 open-ended score) and leading open-source model Qwen3-VL-235B-A22B (49.34 MCQ accuracy, 42.62 open-ended score). However, specialized medical MLLMs do not reliably outperform strong general models, and long-tail generalization emerges as the dominant failure mode. Med-CMR thus provides a stress test for visual-reasoning integration and rare-case robustness in medical MLLMs, and a rigorous yardstick for future clinical systems.

Xinlei Yu, Chengming Xu, Guibin Zhang et al.

Despite the remarkable success of Vision-Language Models (VLMs), their performance on a range of complex visual tasks is often hindered by a "visual processing bottleneck": a propensity to lose grounding in visual evidence and exhibit a deficit in contextualized visual experience during prolonged generation. Drawing inspiration from human cognitive memory theory, which distinguishes short-term visually-dominant memory and long-term semantically-dominant memory, we propose VisMem, a cognitively-aligned framework that equips VLMs with dynamic latent vision memories, a short-term module for fine-grained perceptual retention and a long-term module for abstract semantic consolidation. These memories are seamlessly invoked during inference, allowing VLMs to maintain both perceptual fidelity and semantic consistency across thinking and generation. Extensive experiments across diverse visual benchmarks for understanding, reasoning, and generation reveal that VisMem delivers a significant average performance boost of 11.8% relative to the vanilla model and outperforms all counterparts, establishing a new paradigm for latent-space memory enhancement. The code will be available: https://github.com/YU-deep/VisMem.git.

Xiaofeng Mao, Zhengkai Jiang, Qilin Wang et al.

Recent advancements in the field of Diffusion Transformers have substantially improved the generation of high-quality 2D images, 3D videos, and 3D shapes. However, the effectiveness of the Transformer architecture in the domain of co-speech gesture generation remains relatively unexplored, as prior methodologies have predominantly employed the Convolutional Neural Network (CNNs) or simple a few transformer layers. In an attempt to bridge this research gap, we introduce a novel Masked Diffusion Transformer for co-speech gesture generation, referred to as MDT-A2G, which directly implements the denoising process on gesture sequences. To enhance the contextual reasoning capability of temporally aligned speech-driven gestures, we incorporate a novel Masked Diffusion Transformer. This model employs a mask modeling scheme specifically designed to strengthen temporal relation learning among sequence gestures, thereby expediting the learning process and leading to coherent and realistic motions. Apart from audio, Our MDT-A2G model also integrates multi-modal information, encompassing text, emotion, and identity. Furthermore, we propose an efficient inference strategy that diminishes the denoising computation by leveraging previously calculated results, thereby achieving a speedup with negligible performance degradation. Experimental results demonstrate that MDT-A2G excels in gesture generation, boasting a learning speed that is over 6$\times$ faster than traditional diffusion transformers and an inference speed that is 5.7$\times$ than the standard diffusion model.

Linxuan Fan, Juntao Jiang, Weixuan Liu et al.

Skin lesion segmentation is a crucial step in dermatology for guiding clinical decision-making. However, existing methods for accurate, robust, and resource-efficient lesion analysis have limitations, including low performance and high computational complexity. To address these limitations, we propose UltraLBM-UNet, a lightweight U-Net variant that integrates a bidirectional Mamba-based global modeling mechanism with multi-branch local feature perception. The proposed architecture integrates efficient local feature injection with bidirectional state-space modeling, enabling richer contextual interaction across spatial dimensions while maintaining computational compactness suitable for point-of-care deployment. Extensive experiments on the ISIC 2017, ISIC 2018, and PH2 datasets demonstrate that our model consistently achieves state-of-the-art segmentation accuracy, outperforming existing lightweight and Mamba counterparts with only 0.034M parameters and 0.060 GFLOPs. In addition, we introduce a hybrid knowledge distillation strategy to train an ultra-compact student model, where the distilled variant UltraLBM-UNet-T, with only 0.011M parameters and 0.019 GFLOPs, achieves competitive segmentation performance. These results highlight the suitability of UltraLBM-UNet for point-of-care deployment, where accurate and robust lesion analyses are essential. The source code is publicly available at https://github.com/LinLinLin-X/UltraLBM-UNet.

Yizhang Jin, Jian Li, Jiangning Zhang et al.

Visual Spatial Description (VSD) aims to generate texts that describe the spatial relationships between objects within images. Traditional visual spatial relationship classification (VSRC) methods typically output the spatial relationship between two objects in an image, often neglecting world knowledge and lacking general language capabilities. In this paper, we propose a Large Language-and-Vision Assistant for Visual Spatial Description, named LLaVA-VSD, which is designed for the classification, description, and open-ended description of visual spatial relationships. Specifically, the model first constructs a VSD instruction-following dataset using given figure-caption pairs for the three tasks. It then employs LoRA to fine-tune a Large Language and Vision Assistant for VSD, which has 13 billion parameters and supports high-resolution images. Finally, a large language model (Qwen-2) is used to refine the generated sentences, enhancing their diversity and accuracy. LLaVA-VSD demonstrates excellent multimodal conversational capabilities and can follow open-ended instructions to assist with inquiries about object relationships in images.

Yabiao Wang, Shuo Wang, Jiangning Zhang et al.

Human-human motion generation is essential for understanding humans as social beings. Current methods fall into two main categories: single-person-based methods and separate modeling-based methods. To delve into this field, we abstract the overall generation process into a general framework MetaMotion, which consists of two phases: temporal modeling and interaction mixing. For temporal modeling, the single-person-based methods concatenate two people into a single one directly, while the separate modeling-based methods skip the modeling of interaction sequences. The inadequate modeling described above resulted in sub-optimal performance and redundant model parameters. In this paper, we introduce TIMotion (Temporal and Interactive Modeling), an efficient and effective framework for human-human motion generation. Specifically, we first propose Causal Interactive Injection to model two separate sequences as a causal sequence leveraging the temporal and causal properties. Then we present Role-Evolving Scanning to adjust to the change in the active and passive roles throughout the interaction. Finally, to generate smoother and more rational motion, we design Localized Pattern Amplification to capture short-term motion patterns. Extensive experiments on InterHuman and InterX demonstrate that our method achieves superior performance. Project page: https://aigc-explorer.github.io/TIMotion-page/

Xijie Huang, Chengming Xu, Donghao Luo et al. · tencent-ai

First-Frame Propagation (FFP) offers a promising paradigm for controllable video editing, but existing methods are hampered by a reliance on cumbersome run-time guidance. We identify the root cause of this limitation as the inadequacy of current training datasets, which are often too short, low-resolution, and lack the task diversity required to teach robust temporal priors. To address this foundational data gap, we first introduce FFP-300K, a new large-scale dataset comprising 300K high-fidelity video pairs at 720p resolution and 81 frames in length, constructed via a principled two-track pipeline for diverse local and global edits. Building on this dataset, we propose a novel framework designed for true guidance-free FFP that resolves the critical tension between maintaining first-frame appearance and preserving source video motion. Architecturally, we introduce Adaptive Spatio-Temporal RoPE (AST-RoPE), which dynamically remaps positional encodings to disentangle appearance and motion references. At the objective level, we employ a self-distillation strategy where an identity propagation task acts as a powerful regularizer, ensuring long-term temporal stability and preventing semantic drift. Comprehensive experiments on the EditVerseBench benchmark demonstrate that our method significantly outperforming existing academic and commercial models by receiving about 0.2 PickScore and 0.3 VLM score improvement against these competitors.

Teng Hu, Jiangning Zhang, Ran Yi et al.

In recent years, the development of diffusion models has led to significant progress in image and video generation tasks, with pre-trained models like the Stable Diffusion series playing a crucial role. Inspired by model pruning which lightens large pre-trained models by removing unimportant parameters, we propose a novel model fine-tuning method to make full use of these ineffective parameters and enable the pre-trained model with new task-specified capabilities. In this work, we first investigate the importance of parameters in pre-trained diffusion models, and discover that the smallest 10% to 20% of parameters by absolute values do not contribute to the generation process. Based on this observation, we propose a method termed SaRA that re-utilizes these temporarily ineffective parameters, equating to optimizing a sparse weight matrix to learn the task-specific knowledge. To mitigate overfitting, we propose a nuclear-norm-based low-rank sparse training scheme for efficient fine-tuning. Furthermore, we design a new progressive parameter adjustment strategy to make full use of the re-trained/finetuned parameters. Finally, we propose a novel unstructural backpropagation strategy, which significantly reduces memory costs during fine-tuning. Our method enhances the generative capabilities of pre-trained models in downstream applications and outperforms traditional fine-tuning methods like LoRA in maintaining model's generalization ability. We validate our approach through fine-tuning experiments on SD models, demonstrating significant improvements. SaRA also offers a practical advantage that requires only a single line of code modification for efficient implementation and is seamlessly compatible with existing methods.

Yabiao Wang, Shuo Wang, Jiangning Zhang et al.

This work aims at a challenging task: human action-reaction synthesis, i.e., generating human reactions conditioned on the action sequence of another person. Currently, autoregressive modeling approaches with vector quantization (VQ) have achieved remarkable performance in motion generation tasks. However, VQ has inherent disadvantages, including quantization information loss, low codebook utilization, etc. In addition, while dividing the body into separate units can be beneficial, the computational complexity needs to be considered. Also, the importance of mutual perception among units is often neglected. In this work, we propose MARRS, a novel framework designed to generate coordinated and fine-grained reaction motions using continuous representations. Initially, we present the Unit-distinguished Motion Variational AutoEncoder (UD-VAE), which segments the entire body into distinct body and hand units, encoding each independently. Subsequently, we propose Action-Conditioned Fusion (ACF), which involves randomly masking a subset of reactive tokens and extracting specific information about the body and hands from the active tokens. Furthermore, we introduce Mutual Unit Modulation (MUM) to facilitate interaction between body and hand units by using the information from one unit to adaptively modulate the other. Finally, for the diffusion model, we employ a compact MLP as a noise predictor for each distinct body unit and incorporate the diffusion loss to model the probability distribution of each token. Both quantitative and qualitative results demonstrate that our method achieves superior performance. Project page: https://aigc-explorer.github.io/MARRS/.

Jieyun Bai, Yitong Tang, Zihao Zhou et al.

Accurate segmentation of cervical structures in transvaginal ultrasound (TVS) is critical for assessing the risk of spontaneous preterm birth (PTB), yet the scarcity of labeled data limits the performance of supervised learning approaches. This paper introduces the Fetal Ultrasound Grand Challenge (FUGC), the first benchmark for semi-supervised learning in cervical segmentation, hosted at ISBI 2025. FUGC provides a dataset of 890 TVS images, including 500 training images, 90 validation images, and 300 test images. Methods were evaluated using the Dice Similarity Coefficient (DSC), Hausdorff Distance (HD), and runtime (RT), with a weighted combination of 0.4/0.4/0.2. The challenge attracted 10 teams with 82 participants submitting innovative solutions. The best-performing methods for each individual metric achieved 90.26\% mDSC, 38.88 mHD, and 32.85 ms RT, respectively. FUGC establishes a standardized benchmark for cervical segmentation, demonstrates the efficacy of semi-supervised methods with limited labeled data, and provides a foundation for AI-assisted clinical PTB risk assessment.

Jiangning Zhang, Xuhai Chen, Yabiao Wang et al.

This work studies a challenging and practical issue known as multi-class unsupervised anomaly detection (MUAD). This problem requires only normal images for training while simultaneously testing both normal and anomaly images across multiple classes. Existing reconstruction-based methods typically adopt pyramidal networks as encoders and decoders to obtain multi-resolution features, often involving complex sub-modules with extensive handcraft engineering. In contrast, a plain Vision Transformer (ViT) showcasing a more straightforward architecture has proven effective in multiple domains, including detection and segmentation tasks. It is simpler, more effective, and elegant. Following this spirit, we explore the use of only plain ViT features for MUAD. We first abstract a Meta-AD concept by synthesizing current reconstruction-based methods. Subsequently, we instantiate a novel ViT-based ViTAD structure, designed incrementally from both global and local perspectives. This model provide a strong baseline to facilitate future research. Additionally, this paper uncovers several intriguing findings for further investigation. Finally, we comprehensively and fairly benchmark various approaches using eight metrics. Utilizing a basic training regimen with only an MSE loss, ViTAD achieves state-of-the-art results and efficiency on MVTec AD, VisA, and Uni-Medical datasets. \Eg, achieving 85.4 mAD that surpasses UniAD by +3.0 for the MVTec AD dataset, and it requires only 1.1 hours and 2.3G GPU memory to complete model training on a single V100 that can serve as a strong baseline to facilitate the development of future research. Full code is available at https://zhangzjn.github.io/projects/ViTAD/.

Yuxuan Cai, Jiangning Zhang, Haoyang He et al.

The success of Large Language Models (LLMs) has inspired the development of Multimodal Large Language Models (MLLMs) for unified understanding of vision and language. However, the increasing model size and computational complexity of large-scale MLLMs (l-MLLMs) limit their use in resource-constrained scenarios. Although small-scale MLLMs (s-MLLMs) are designed to reduce computational costs, they typically suffer from performance degradation. To mitigate this limitation, we propose a novel LLaVA-KD framework to transfer knowledge from l-MLLMs to s-MLLMs. Specifically, we introduce Multimodal Distillation (MDist) to transfer teacher model's robust representations across both visual and linguistic modalities, and Relation Distillation (RDist) to transfer teacher model's ability to capture visual token relationships. Additionally, we propose a three-stage training scheme to fully exploit the potential of the proposed distillation strategy: 1) Distilled Pre-Training to strengthen the alignment between visual-linguistic representations in s-MLLMs, 2) Supervised Fine-Tuning to equip the s-MLLMs with multimodal understanding capacity, and 3) Distilled Fine-Tuning to refine s-MLLM's knowledge. Our approach significantly improves s-MLLMs performance without altering the model architecture. Extensive experiments and ablation studies validate the effectiveness of each proposed component. Code will be available at https://github.com/Fantasyele/LLaVA-KD.

Yue Han, Jiangning Zhang, Junwei Zhu et al.

This work presents FaceX framework, a novel facial generalist model capable of handling diverse facial tasks simultaneously. To achieve this goal, we initially formulate a unified facial representation for a broad spectrum of facial editing tasks, which macroscopically decomposes a face into fundamental identity, intra-personal variation, and environmental factors. Based on this, we introduce Facial Omni-Representation Decomposing (FORD) for seamless manipulation of various facial components, microscopically decomposing the core aspects of most facial editing tasks. Furthermore, by leveraging the prior of a pretrained StableDiffusion (SD) to enhance generation quality and accelerate training, we design Facial Omni-Representation Steering (FORS) to first assemble unified facial representations and then effectively steer the SD-aware generation process by the efficient Facial Representation Controller (FRC). %Without any additional features, Our versatile FaceX achieves competitive performance compared to elaborate task-specific models on popular facial editing tasks. Full codes and models will be available at https://github.com/diffusion-facex/FaceX.

Jiangning Zhang, Junwei Zhu, Teng Hu et al.

Native 4K (2160$\times$3840) video generation remains a critical challenge due to the quadratic computational explosion of full-attention as spatiotemporal resolution increases, making it difficult for models to strike a balance between efficiency and quality. This paper proposes a novel Transformer retrofit strategy termed $\textbf{T3}$ ($\textbf{T}$ransform $\textbf{T}$rained $\textbf{T}$ransformer) that, without altering the core architecture of full-attention pretrained models, significantly reduces compute requirements by optimizing their forward logic. Specifically, $\textbf{T3-Video}$ introduces a multi-scale weight-sharing window attention mechanism and, via hierarchical blocking together with an axis-preserving full-attention design, can effect an "attention pattern" transformation of a pretrained model using only modest compute and data. Results on 4K-VBench show that $\textbf{T3-Video}$ substantially outperforms existing approaches: while delivering performance improvements (+4.29$\uparrow$ VQA and +0.08$\uparrow$ VTC), it accelerates native 4K video generation by more than 10$\times$. Project page at https://zhangzjn.github.io/projects/T3-Video

Jiangning Zhang, Chengjie Wang, Xiangtai Li et al.

Anomaly detection (AD) is often focused on detecting anomaly areas for industrial quality inspection and medical lesion examination. However, due to the specific scenario targets, the data scale for AD is relatively small, and evaluation metrics are still deficient compared to classic vision tasks, such as object detection and semantic segmentation. To fill these gaps, this work first constructs a large-scale and general-purpose COCO-AD dataset by extending COCO to the AD field. This enables fair evaluation and sustainable development for different methods on this challenging benchmark. Moreover, current metrics such as AU-ROC have nearly reached saturation on simple datasets, which prevents a comprehensive evaluation of different methods. Inspired by the metrics in the segmentation field, we further propose several more practical threshold-dependent AD-specific metrics, ie, m$F_1$$^{.2}_{.8}$, mAcc$^{.2}_{.8}$, mIoU$^{.2}_{.8}$, and mIoU-max. Motivated by GAN inversion's high-quality reconstruction capability, we propose a simple but more powerful InvAD framework to achieve high-quality feature reconstruction. Our method improves the effectiveness of reconstruction-based methods on popular MVTec AD, VisA, and our newly proposed COCO-AD datasets under a multi-class unsupervised setting, where only a single detection model is trained to detect anomalies from different classes. Extensive ablation experiments have demonstrated the effectiveness of each component of our InvAD. Full codes and models are available at https://github.com/zhangzjn/ader.

Yongqi Yang, Huayang Huang, Xu Peng et al.

Recent hybrid video generation models combine autoregressive temporal dynamics with diffusion-based spatial denoising, but their sequential, iterative nature leads to error accumulation and long inference times. In this work, we propose a distillation-based framework for efficient causal video generation that enables high-quality synthesis with extremely limited denoising steps. Our approach builds upon the Distribution Matching Distillation (DMD) framework and proposes a novel Adversarial Self-Distillation (ASD) strategy, which aligns the outputs of the student model's n-step denoising process with its (n+1)-step version at the distribution level. This design provides smoother supervision by bridging small intra-student gaps and more informative guidance by combining teacher knowledge with locally consistent student behavior, substantially improving training stability and generation quality in extremely few-step scenarios (e.g., 1-2 steps). In addition, we present a First-Frame Enhancement (FFE) strategy, which allocates more denoising steps to the initial frames to mitigate error propagation while applying larger skipping steps to later frames. Extensive experiments on VBench demonstrate that our method surpasses state-of-the-art approaches in both one-step and two-step video generation. Notably, our framework produces a single distilled model that flexibly supports multiple inference-step settings, eliminating the need for repeated re-distillation and enabling efficient, high-quality video synthesis.

Wenbing Zhu, Chengjie Wang, Bin-Bin Gao et al.

Industrial Anomaly Detection (IAD) is critical for enhancing operational safety, ensuring product quality, and optimizing manufacturing efficiency across global industries. However, the IAD algorithms are severely constrained by the limitations of existing public benchmarks. Current datasets exhibit restricted category diversity and insufficient scale, frequently resulting in metric saturation and limited model transferability to real-world scenarios. To address this gap, we introduce Real-IAD Variety, the largest and most diverse IAD benchmark, comprising 198,960 high-resolution images across 160 distinct object categories. Its diversity is ensured through comprehensive coverage of 28 industries, 24 material types, and 22 color variations. Our comprehensive experimental analysis validates the benchmark's substantial challenge: state-of-the-art multi-class unsupervised anomaly detection methods experience significant performance degradation when scaled from 30 to 160 categories. Crucially, we demonstrate that vision-language models exhibit remarkable robustness to category scale-up, with minimal performance variation across different category counts, significantly enhancing generalization capabilities in diverse industrial contexts. The unprecedented scale and complexity of Real-IAD Variety position it as an essential resource for training and evaluating next-generation foundation models for anomaly detection. By providing this comprehensive benchmark with rigorous evaluation protocols across multi-class unsupervised, multi-view, and zero-/few-shot settings, we aim to accelerate research beyond domain-specific constraints, enabling the development of scalable, general-purpose anomaly detection systems. Real-IAD Variety will be made publicly available to facilitate innovation in this critical field.

Zhucun Xue, Jiangning Zhang, Teng Hu et al.

The quality of the video dataset (image quality, resolution, and fine-grained caption) greatly influences the performance of the video generation model. The growing demand for video applications sets higher requirements for high-quality video generation models. For example, the generation of movie-level Ultra-High Definition (UHD) videos and the creation of 4K short video content. However, the existing public datasets cannot support related research and applications. In this paper, we first propose a high-quality open-sourced UHD-4K (22.4\% of which are 8K) text-to-video dataset named UltraVideo, which contains a wide range of topics (more than 100 kinds), and each video has 9 structured captions with one summarized caption (average of 824 words). Specifically, we carefully design a highly automated curation process with four stages to obtain the final high-quality dataset: \textit{i)} collection of diverse and high-quality video clips. \textit{ii)} statistical data filtering. \textit{iii)} model-based data purification. \textit{iv)} generation of comprehensive, structured captions. In addition, we expand Wan to UltraWan-1K/-4K, which can natively generate high-quality 1K/4K videos with more consistent text controllability, demonstrating the effectiveness of our data curation.We believe that this work can make a significant contribution to future research on UHD video generation. UltraVideo dataset and UltraWan models are available at https://xzc-zju.github.io/projects/UltraVideo.

Yuhu Bai, Jiangning Zhang, Zhaofeng Chen et al.

Few-shot anomaly detection (FSAD) plays a crucial role in industrial manufacturing. However, existing FSAD methods encounter difficulties leveraging a limited number of normal samples, frequently failing to detect and locate inconspicuous anomalies in the spatial domain. We have further discovered that these subtle anomalies would be more noticeable in the frequency domain. In this paper, we propose a Dual-Path Frequency Discriminators (DFD) network from a frequency perspective to tackle these issues. The original spatial images are transformed into multi-frequency images, making them more conducive to the tailored discriminators in detecting anomalies. Additionally, the discriminators learn a joint representation with forms of pseudo-anomalies. Extensive experiments conducted on MVTec AD and VisA benchmarks demonstrate that our DFD surpasses current state-of-the-art methods. The code is available at \url{https://github.com/yuhbai/DFD}.

Teng Hu, Jiangning Zhang, Ran Yi et al.

Employing LLMs for visual generation has recently become a research focus. However, the existing methods primarily transfer the LLM architecture to visual generation but rarely investigate the fundamental differences between language and vision. This oversight may lead to suboptimal utilization of visual generation capabilities within the LLM framework. In this paper, we explore the characteristics of visual embedding space under the LLM framework and discover that the correlation between visual embeddings can help achieve more stable and robust generation results. We present IAR, an Improved AutoRegressive Visual Generation Method that enhances the training efficiency and generation quality of LLM-based visual generation models. Firstly, we propose a Codebook Rearrangement strategy that uses balanced k-means clustering algorithm to rearrange the visual codebook into clusters, ensuring high similarity among visual features within each cluster. Leveraging the rearranged codebook, we propose a Cluster-oriented Cross-entropy Loss that guides the model to correctly predict the cluster where the token is located. This approach ensures that even if the model predicts the wrong token index, there is a high probability the predicted token is located in the correct cluster, which significantly enhances the generation quality and robustness. Extensive experiments demonstrate that our method consistently enhances the model training efficiency and performance from 100M to 1.4B, reducing the training time by half while achieving the same FID. Additionally, our approach can be applied to various LLM-based visual generation models and adheres to the scaling law, providing a promising direction for future research in LLM-based visual generation. The code is available at: https://github.com/sjtuplayer/IAR.

Ke Fan, Jiangning Zhang, Ran Yi et al.

Text-to-motion generation is a crucial task in computer vision, which generates the target 3D motion by the given text. The existing annotated datasets are limited in scale, resulting in most existing methods overfitting to the small datasets and unable to generalize to the motions of the open domain. Some methods attempt to solve the open-vocabulary motion generation problem by aligning to the CLIP space or using the Pretrain-then-Finetuning paradigm. However, the current annotated dataset's limited scale only allows them to achieve mapping from sub-text-space to sub-motion-space, instead of mapping between full-text-space and full-motion-space (full mapping), which is the key to attaining open-vocabulary motion generation. To this end, this paper proposes to leverage the atomic motion (simple body part motions over a short time period) as an intermediate representation, and leverage two orderly coupled steps, i.e., Textual Decomposition and Sub-motion-space Scattering, to address the full mapping problem. For Textual Decomposition, we design a fine-grained description conversion algorithm, and combine it with the generalization ability of a large language model to convert any given motion text into atomic texts. Sub-motion-space Scattering learns the compositional process from atomic motions to the target motions, to make the learned sub-motion-space scattered to form the full-motion-space. For a given motion of the open domain, it transforms the extrapolation into interpolation and thereby significantly improves generalization. Our network, $DSO$-Net, combines textual $d$ecomposition and sub-motion-space $s$cattering to solve the $o$pen-vocabulary motion generation. Extensive experiments demonstrate that our DSO-Net achieves significant improvements over the state-of-the-art methods on open-vocabulary motion generation. Code is available at https://vankouf.github.io/DSONet/.

Zhucun Xue, Jiangning Zhang, Xurong Xie et al.

Multimodal Large Language Models (MLLMs) struggle with long videos due to fixed context windows and weak long-term dependency modeling. Existing Retrieval-Augmented Generation (RAG) methods for videos use static retrieval strategies, leading to inefficiencies for simple queries and information loss for complex tasks. To address this, we propose AdaVideoRAG, a novel framework that dynamically adapts retrieval granularity based on query complexity using a lightweight intent classifier. Our framework employs an Omni-Knowledge Indexing module to build hierarchical databases from text (captions, ASR, OCR), visual features, and semantic graphs, enabling optimal resource allocation across tasks. We also introduce the HiVU benchmark for comprehensive evaluation. Experiments demonstrate improved efficiency and accuracy for long-video understanding, with seamless integration into existing MLLMs. AdaVideoRAG establishes a new paradigm for adaptive retrieval in video analysis. Codes will be open-sourced at https://github.com/xzc-zju/AdaVideoRAG.

Teng Hu, Zhentao Yu, Zhengguang Zhou et al.

Despite recent advances in video generation, existing models still lack fine-grained controllability, especially for multi-subject customization with consistent identity and interaction. In this paper, we propose PolyVivid, a multi-subject video customization framework that enables flexible and identity-consistent generation. To establish accurate correspondences between subject images and textual entities, we design a VLLM-based text-image fusion module that embeds visual identities into the textual space for precise grounding. To further enhance identity preservation and subject interaction, we propose a 3D-RoPE-based enhancement module that enables structured bidirectional fusion between text and image embeddings. Moreover, we develop an attention-inherited identity injection module to effectively inject fused identity features into the video generation process, mitigating identity drift. Finally, we construct an MLLM-based data pipeline that combines MLLM-based grounding, segmentation, and a clique-based subject consolidation strategy to produce high-quality multi-subject data, effectively enhancing subject distinction and reducing ambiguity in downstream video generation. Extensive experiments demonstrate that PolyVivid achieves superior performance in identity fidelity, video realism, and subject alignment, outperforming existing open-source and commercial baselines.

Yuji Wang, Moran Li, Xiaobin Hu et al.

Identity-preserving text-to-video (IPT2V) generation, which aims to create high-fidelity videos with consistent human identity, has become crucial for downstream applications. However, current end-to-end frameworks suffer a critical spatial-temporal trade-off: optimizing for spatially coherent layouts of key elements (e.g., character identity preservation) often compromises instruction-compliant temporal smoothness, while prioritizing dynamic realism risks disrupting the spatial coherence of visual structures. To tackle this issue, we propose a simple yet effective spatial-temporal decoupled framework that decomposes representations into spatial features for layouts and temporal features for motion dynamics. Specifically, our paper proposes a semantic prompt optimization mechanism and stage-wise decoupled generation paradigm. The former module decouples the prompt into spatial and temporal components. Aligned with the subsequent stage-wise decoupled approach, the spatial prompts guide the text-to-image (T2I) stage to generate coherent spatial features, while the temporal prompts direct the sequential image-to-video (I2V) stage to ensure motion consistency. Experimental results validate that our approach achieves excellent spatiotemporal consistency, demonstrating outstanding performance in identity preservation, text relevance, and video quality. By leveraging this simple yet robust mechanism, our algorithm secures the runner-up position in 2025 ACM MultiMedia Challenge. Our code is available at https://github.com/rain152/IPVG.

Shijun Shi, Jing Xu, Zhihang Li et al.

Recent advances in diffusion models have greatly improved pose-driven character animation. However, existing methods are limited to spatially aligned reference-pose pairs with matched skeletal structures. Handling reference-pose misalignment remains unsolved. To address this, we present One-to-All Animation, a unified framework for high-fidelity character animation and image pose transfer for references with arbitrary layouts. First, to handle spatially misaligned reference, we reformulate training as a self-supervised outpainting task that transforms diverse-layout reference into a unified occluded-input format. Second, to process partially visible reference, we design a reference extractor for comprehensive identity feature extraction. Further, we integrate hybrid reference fusion attention to handle varying resolutions and dynamic sequence lengths. Finally, from the perspective of generation quality, we introduce identity-robust pose control that decouples appearance from skeletal structure to mitigate pose overfitting, and a token replace strategy for coherent long-video generation. Extensive experiments show that our method outperforms existing approaches. The code and model are available at https://github.com/ssj9596/One-to-All-Animation.

Haiyue Sun, Qingdong He, Jinlong Peng et al.

Autoregressive Model (AR) has shown remarkable success in conditional image generation. However, these approaches for multiple reference generation struggle with decoupling different reference identities. In this work, we propose the TokenAR framework, specifically focused on a simple but effective token-level enhancement mechanism to address reference identity confusion problem. Such token-level enhancement consists of three parts, 1). Token Index Embedding clusters the tokens index for better representing the same reference images; 2). Instruct Token Injection plays as a role of extra visual feature container to inject detailed and complementary priors for reference tokens; 3). The identity-token disentanglement strategy (ITD) explicitly guides the token representations toward independently representing the features of each identity.This token-enhancement framework significantly augments the capabilities of existing AR based methods in conditional image generation, enabling good identity consistency while preserving high quality background reconstruction. Driven by the goal of high-quality and high-diversity in multi-subject generation, we introduce the InstructAR Dataset, the first open-source, large-scale, multi-reference input, open domain image generation dataset that includes 28K training pairs, each example has two reference subjects, a relative prompt and a background with mask annotation, curated for multiple reference image generation training and evaluating. Comprehensive experiments validate that our approach surpasses current state-of-the-art models in multiple reference image generation task. The implementation code and datasets will be made publicly. Codes are available, see https://github.com/lyrig/TokenAR