论文检索

Huangbiao Xu, Xiao Ke, Huanqi Wu, Rui Xu, Yuezhou Li, Peirong Xu, Wenzhong Guo

The fair and objective assessment of performances and competitions is a common pursuit and challenge in human society. The application of computer vision technology offers hope for this purpose, but it still faces obstacles such as occlusion and motion blur. To address these hindrances, our DanceFix proposes a bidirectional spatial-temporal context optical flow correction (BOFC) method. This approach leverages the consistency and complementarity of motion information between two modalities: optical flow, which excels at pixel capture, and lightweight skeleton data. It enables the extraction of pixel-level motion changes and the correction of abnormal skeleton data. Furthermore, we propose a part-level dance dataset (Dancer Parts) and part-level motion feature extraction based on task decoupling (PETD). This aims to decouple complex whole-body parts tracking into fine-grained limb-level motion extraction, enhancing the confidence of temporal information and the accuracy of correction for abnormal data. Finally, we present the DNV dataset, which simulates fully neat group dance scenes and provides reliable labels and validation methods for the newly introduced group dance neatness assessment (GDNA). To the best of our knowledge, this is the first work to develop quantitative criteria for assessing limb and joint neatness in group dance. We conduct experiments on DNV and video-based public JHMDB datasets. Our method effectively corrects abnormal skeleton points, flexibly embeds, and improves the accuracy of existing pose estimation algorithms.

Haoran Xu, Peixi Peng, Xinyi Zhang, Guang Tan, Yaokun Li, Shuaixian Wang, Luntong Li

Occupancy networks aim to reconstruct the surroundings with occupied semantic voxels. However, frequent object occlusions often occur in dynamic real-world scenarios, which cannot be captured by independent frames. Most existing occupancy networks generate results without explicitly considering past occupancy states and continuous visual changes over time, limiting their temporal accuracy. We tackle it by treating the task from a new continuous updating perspective, which considers historical data and continuous motion clues. We propose a new approach termed Continuous Motion clue exploitation for Occupancy Prediction (CMOP), which incorporates three key designs: (i) Propagator: which forecasts future occupancy states based on historical data; (ii) Tracker: which updates the occupancy on a per-frame basis using dynamic visual motion information; and (iii) Fuser: which aggregates results from the Propagator and Tracker into more robust and accurate occupancy results. Experiments on several benchmarks demonstrate that CMOP outperforms state-of-the-art baselines.

Hai-Ming Xu, Qi Chen, Lei Wang, Lingqiao Liu

Recent advancements in Multimodal Large Language Models (MLLMs) have generated significant interest in their ability to autonomously interact with and interpret Graphical User Interfaces (GUIs). A major challenge in these systems is grounding—accurately identifying critical GUI components such as text or icons based on a GUI image and a corresponding text query. Traditionally, this task has relied on fine-tuning MLLMs with specialized training data to predict component locations directly. However, in this paper, we propose a novel Tuning-free Attention-driven Grounding (TAG) method that leverages the inherent attention patterns in pretrained MLLMs to accomplish this task without the need for additional fine-tuning. Our method involves identifying and aggregating attention maps from specific tokens within a carefully constructed query prompt. Applied to MiniCPM-Llama3-V 2.5, a state-of-the-art MLLM, our tuning-free approach achieves performance comparable to tuning-based methods, with notable success in text localization. Additionally, we demonstrate that our attention map-based grounding technique significantly outperforms direct localization predictions from MiniCPM-Llama3-V 2.5, highlighting the potential of using attention maps from pretrained MLLMs and paving the way for future innovations in this domain.

Feilong Xu, Feiyang Yang, Xiongfei Li, Xiaoli Zhang

In utilizing deep learning techniques for medical image segmentation, two types of imbalance issues are observed: inter-class imbalance between majority and minority classes and intra-class imbalance between easy and hard samples. However, existing loss functions typically confuse these issues, leading to enhancements that cater to only one aspect. Moreover, loss functions optimized for specific tasks often exhibit limited generalizability. To address these issues, we propose Inter-class and Intra-class Balance loss, as well as a unified loss termed Balance loss. The Inter-class Balance loss controls the extent of hard sample mining for majority class samples by considering the frequency of minority classes present in each input image. This approach requires no manual adjustment weights and adapts automatically to different datasets. The Intra-class Balance loss enhances the network's ability to learn from hard samples by performing mining on hard samples within each class. We evaluate our loss functions on five segmentation tasks with varying degrees of class imbalance. The experimental results show that our proposed Balance loss enhances segmentation performance compared with the current loss functions and exhibits superior robustness.

FeiFan Xu, Tianyi Chen, Fan Yang, Yunfei Zhang, Si Wu

The rapid advancement of 3D Generative Adversarial Networks (GANs) has significantly enhanced the diversity and quality of generated 3D images. Despite these breakthroughs, the manipulation capabilities of 3D GANs remain unexplored, presenting substantial challenges for practical applications where user interaction and modification are essential. Current manipulation methods often lack the precision needed for fine-grained attribute manipulation, and struggle to maintain multi-view consistency during the editing process. To address these limitations, we propose 3DHumanEdit, a novel approach for 3D human body part-aware manipulation. 3DHumanEdit leverages multi-modal feature fusion and body part-aware feature alignment to achieve precise manipulation of individual body parts based on detailed text inputs and segmentation images. By exploring 3D prior for accurate editing and enforcing correspondence in latent space, 3DHumanEdit ensures coherence across multiple views. Experiments demonstrate that 3DHumanEdit outperforms existing methods in both editing fidelity and multi-view consistency, offering a robust solution for fine-grained 3D manipulation.

Chenlong Xu, Bineng Zhong, Qihua Liang, Yaozong Zheng, Guorong Li, Shuxiang Song

Recently, several studies have shown that utilizing contextual information to perceive target states is crucial for object tracking. They typically capture context by incorporating multiple video frames. However, these naive frame-context methods fail to consider the importance of each patch within a reference frame, making them susceptible to noise and redundant tokens, which deteriorates tracking performance. To address this challenge, we propose a new token context-aware tracking pipeline named LMTrack, designed to automatically learn high-quality reference tokens for efficient visual tracking. Embracing the principle of Less is More, the core idea of LMTrack is to analyze the importance distribution of all reference tokens, where important tokens are collected, continually attended to, and updated. Specifically, a novel Token Context Memory module is designed to dynamically collect high-quality spatio-temporal information of a target in an autoregressive manner, eliminating redundant background tokens from the reference frames. Furthermore, an effective Unidirectional Token Attention mechanism is designed to establish dependencies between reference tokens and search frame, enabling robust cross-frame association and target localization. Extensive experiments demonstrate the superiority of our tracker, achieving state-of-the-art results on tracking benchmarks such as GOT-10K, TrackingNet, and LaSOT.

Cheng Xu, Weiwen Zhang, Hongrui Zhang, Xuemiao Xu, Huaidong Zhang, Jing Zou, Jing Qin

In clinical imaging, medical segmentation networks typically require continually adapting to new data from multiple sites over time, as aggregating all data for learning at once can be impractical due to storage limitations and privacy concerns. However, existing methods basically overlook domain-specific characteristics and fall short of adequately capturing domain-invariant knowledge during continual learning, leading to undesired catastrophic forgetting of previous sites and inferior generalization to new sites. To tackle this issue, this paper introduces FR2Seg, to sufficiently exploit both domain-specific and domain-invariant knowledge for efficient continual learning with the aid of low-frequency cues. For the former aspect, we propose a Fourier style replay module to synthesize pseudo images with old-site styles for data augmentation during new-site training, effectively preventing catastrophic forgetting without sacrificing data privacy. For the latter, we present a Fourier adaptive consistency regularization to identify and constrain the optimization of domain-invariant parameters with explicit awareness of knowledge transferability across sites, ensuring excellent generalizability to new sites. Experimental results on two public datasets confirm our method's superiority over existing state-of-the-art continual learning methods.

Cheng Xu, Fei Hou, Wencheng Wang, Hong Qin, Zhebin Zhang, Ying He

While Signed Distance Fields (SDF) are well-established for modeling watertight surfaces, Unsigned Distance Fields (UDF) broaden the scope to include open surfaces and models with complex inner structures. Despite their flexibility, UDFs encounter significant challenges in high-fidelity 3D reconstruction, such as non-differentiability at the zero level set, difficulty in achieving the exact zero value, numerous local minima, vanishing gradients, and oscillating gradient directions near the zero level set. To address these challenges, we propose Details Enhanced UDF (DEUDF) learning that integrates normal alignment and the SIREN network for capturing fine geometric details, adaptively weighted Eikonal constraints to address vanishing gradients near the target surface, unconditioned MLP-based UDF representation to relax non-negativity constraints, and DCUDF for extracting the local minimal average distance surface. These strategies collectively stabilize the learning process from unoriented point clouds and enhance the accuracy of UDFs. Our computational results demonstrate that DEUDF outperforms existing UDF learning methods in both accuracy and the quality of reconstructed surfaces.

Chejian Xu, Aleksandr Petiushko, Ding Zhao, Bo Li

The field of Autonomous Driving (AD) has witnessed significant progress in recent years. Among the various challenges faced, the safety evaluation of autonomous vehicles (AVs) stands out as a critical concern. Traditional evaluation methods are both costly and inefficient, often requiring extensive driving mileage in order to encounter rare safety-critical scenarios, which are distributed on the long tail of the complex real-world driving landscape. In this paper, we propose a unified approach, Diffusion-Based Safety-Critical Scenario Generation (DiffScene), to generate high-quality safety-critical scenarios which are both realistic and safety-critical for efficient AV evaluation. In particular, we propose a diffusion-based generation framework, leveraging the power of approximating the distribution of low-density spaces for diffusion models. We design several adversarial optimization objectives to guide the diffusion generation under predefined adversarial budgets. These objectives, such as safety-based objective, functionality-based objective, and constraint-based objective, ensure the generation of safety-critical scenarios while adhering to specific constraints. Extensive experimentation has been conducted to validate the efficacy of our approach. Compared with 6 SOTA baselines, DiffScene generates scenarios that are (1) more safety-critical under 3 metrics, (2) more realistic under 5 distance functions, and (3) more transferable to different AV algorithms. In addition, we demonstrate that training AV algorithms with scenarios generated by DiffScene leads to significantly higher performance in terms of the safety-critical metrics compared to baselines. These findings highlight the potential of DiffScene in addressing the challenges of AV safety evaluation, paving the way for safer AV development.

Jingqiao Xiu, Mengze Li, Zongxin Yang, Wei Ji, Yifang Yin, Roger Zimmermann

Audio-Visual Semantic Segmentation (AVSS) has gained significant attention in the multi-modal domain, aiming to segment video objects that produce specific sounds in the corresponding audio. Despite notable progress, existing methods still struggle to handle new classes not included in the original training set. To this end, we introduce Few-Shot Incremental Learning (FSIL) to the AVSS task, which seeks to seamlessly integrate new classes with limited incremental samples while preserving the knowledge of old classes. Two challenges arise in this new setting: (1) To reduce labeling costs, old classes within the incremental samples are treated as background, similar to silent objects. Training the model directly with background annotations may worsen the loss of distinctive knowledge about old classes, such as their outlines and sounds. (2) Most existing models adopt early cross-modal fusion with a single-tower design, incorporating more characteristics into class representations, which impedes knowledge transfer between classes based on similarity. To address these issues, we propose a Few-shot Incremental learning framework via class-centric foregrouNd aggreGation and dual-tower knowlEdge tRansfer (FINGER) for the AVSS task, which comprises two targeted modules: (1) The class-centric foreground aggregation gathers class-specific features for each foreground class while disregarding background features. The background class is excluded during training and inferred from the foreground predictions. (2) The dual-tower knowledge transfer postpones cross-modal fusion to separately conduct knowledge transfer for each modality. Extensive experiments validate the effectiveness of the FINGER model, significantly surpassing state-of-the-art methods.

Bing Xiong, Yue Peng, Ranran Zhang, Fuqiang Chen, Jiaye He, Wenjian Qin

Virtual staining leverages computer-aided techniques to transfer the style of histochemically stained tissue samples to other staining types. In virtual staining of pathological images, maintaining strict structural consistency is crucial, as these images emphasize structural integrity more than natural images. Even slight structural alterations can lead to deviations in diagnostic semantic information. Furthermore, the unpaired characteristic of virtual staining data may compromise the preservation of pathological diagnostic content. To address these challenges, we propose a dual-path inversion virtual staining method using prompt learning, which optimizes visual prompts to control content and style, while preserving complete pathological diagnostic content. Our proposed inversion technique comprises two key components: (1) Dual Path Prompted Strategy, we utilize a feature adapter function to generate reference images for inversion, providing style templates for input image inversion, called Style Target Path. We utilize the inversion of the input image as the Structural Target path, employing visual prompt images to maintain structural consistency in this path while preserving style information from the style Target path. During the deterministic sampling process, we achieve complete content-style disentanglement through a plug-and-play embedding visual prompt approach. (2) StainPrompt Optimization, where we only optimize the null visual prompt as ``operator'' for dual path inversion, rather than fine-tune pre-trained model. We optimize null visual prompt for structual and style trajectory around pivotal noise on each timestep, ensuring accurate dual-path inversion reconstruction. Extensive evaluations on publicly available multi-domain unpaired staining datasets demonstrate high structural consistency and accurate style transfer results.

Zhuyang Xie, Yan Yang, Yankai Yu, Jie Wang, Yongquan Jiang, Xiao Wu

Dense video captioning aims to detect and describe all events in untrimmed videos. This paper presents a dense video captioning network called Multi-Concept Cyclic Learning (MCCL), which aims to: (1) detect multiple concepts at the frame level and leverage these concepts to provide temporal event cues; and (2) establish cyclic co-learning between the generator and the localizer within the captioning network to promote semantic perception and event localization. Specifically, weakly supervised concept detection is performed for each frame, and the detected concept embeddings are integrated into the video features to provide event cues. Additionally, video-level concept contrastive learning is introduced to produce more discriminative concept embeddings. In the captioning network, a cyclic co-learning strategy is proposed, where the generator guides the localizer for event localization through semantic matching, while the localizer enhances the generator’s event semantic perception through location matching, making semantic perception and event localization mutually beneficial. MCCL achieves state-of-the-art performance on the ActivityNet Captions and YouCook2 datasets. Extensive experiments demonstrate its effectiveness and interpretability.

Zhifeng Xie, Hao Li, Huiming Ding, Mengtian Li, Xinhan Di, Ying Cao

Fashion design is a challenging and complex process. Recent works on fashion generation and editing are all agnostic of the actual fashion design process, which limits their usage in practice. In this paper, we propose a novel hierarchical diffusion-based framework tailored for fashion design, coined as HieraFashDiff. Our model is designed to mimic the practical fashion design workflow, by unraveling the denosing process into two successive stages: 1) an ideation stage that generates design proposals given high-level concepts and 2) an iteration stage that continuously refines the proposals using low-level attributes. Our model supports fashion design generation and fine-grained local editing in a single framework. To train our model, we contribute a new dataset of full-body fashion images annotated with hierarchical text descriptions. Extensive evaluations show that, as compared to prior approaches, our method can generate fashion designs and edited results with higher fidelity and better prompt adherence, showing its promising potential to augment the practical fashion design workflow.

Yifan Xie, Tao Feng, Xin Zhang, Xiangyang Luo, Zixuan Guo, Weijiang Yu, Heng Chang, Fei Ma, Fei Richard Yu

Talking head synthesis with arbitrary speech audio is a crucial challenge in the field of digital humans. Recently, methods based on radiance fields have received increasing attention due to their ability to synthesize high-fidelity and identity-consistent talking heads from just a few minutes of training video. However, due to the limited scale of the training data, these methods often exhibit poor performance in audio-lip synchronization and visual quality. In this paper, we propose a novel 3D Gaussian-based method called PointTalk, which constructs a static 3D Gaussian field of the head and deforms it in sync with the audio. It also incorporates an audio-driven dynamic lip point cloud as a critical component of the conditional information, thereby facilitating the effective synthesis of talking heads. Specifically, the initial step involves generating the corresponding lip point cloud from the audio signal and capturing its topological structure. The design of the dynamic difference encoder aims to capture the subtle nuances inherent in dynamic lip movements more effectively. Furthermore, we integrate the audio-point enhancement module, which not only ensures the synchronization of the audio signal with the corresponding lip point cloud within the feature space, but also facilitates a deeper understanding of the interrelations among cross-modal conditional features. Extensive experiments demonstrate that our method achieves superior high-fidelity and audio-lip synchronization in talking head synthesis compared to previous methods.

Peijin Xie, Lin Sun, Bingquan Liu, Dexin Wang, Xiangzheng Zhang, Chengjie Sun, Jiajia Zhang

Distinguishing spatial relations is a basic part of human cognition which requires fine-grained perception on cross-instance. Although benchmarks like MME, MMBench and SEED comprehensively have evaluated various capabilities which already include visual spatial reasoning(VSR). There is still a lack of sufficient quantity and quality evaluation and optimization datasets for Vision Large Language Models(VLLMs) specifically targeting visual positional reasoning. To handle this, we first diagnosed current VLLMs with the VSR dataset and proposed a unified test set. We found current VLLMs to exhibit a contradiction of over-sensitivity to language instructions and under-sensitivity to visual positional information. By expanding the original benchmark from two aspects of tunning data and model structure, we mitigated this phenomenon. To our knowledge, we expanded spatially positioned image data controllably using diffusion models for the first time and integrated original visual encoding(CLIP) with other 3 powerful visual encoders(SigLIP, SAM and DINO). After conducting combination experiments on scaling data and models, we obtained a VLLM VSR Expert(VSRE) that not only generalizes better to different instructions but also accurately distinguishes differences in visual positional information. VSRE achieved over a 27% increase in accuracy on the VSR test set. It becomes a performant VLLM on the position reasoning of both the VSR dataset and relevant subsets of other evaluation benchmarks. We hope it will accelerate advancements in VLLM on VSR learning.

Lianxin Xie, Bingbing Zheng, Wen Xue, Yunfei Zhang, Le Jiang, Ruotao Xu, Si Wu, Hau-San Wong

Blind face video restoration aims to restore high-fidelity details from videos subjected to complex and unknown degradations. This task poses a significant challenge of managing temporal heterogeneity while at the same time maintaining stable face attributes. In this paper, we introduce a Discrete Prior-based Temporal-Coherent content prediction transformer to address the challenge, and our model is referred to as DP-TempCoh. Specifically, we incorporate a spatial-temporal-aware content prediction module to synthesize high-quality content from discrete visual priors, conditioned on degraded video tokens. To further enhance the temporal coherence of the predicted content, a motion statistics modulation module is designed to adjust the content, based on discrete motion priors in terms of cross-frame mean and variance. As a result, the statistics of the predicted content can match with that of real videos over time. By performing extensive experiments, we verify the effectiveness of the design elements and demonstrate the superior performance of our DP-TempCoh in both synthetically and naturally degraded video restoration.

Jinxia Xie, Bineng Zhong, Qihua Liang, Ning Li, Zhiyi Mo, Shuxiang Song

How to make a good trade-off between performance and computational cost is crucial for a tracker. However, current famous methods typically focus on complicated and time-consuming learning that combining temporal and appearance information by input more and more images (or features). Consequently, these methods not only increase the model's computational source and learning burden but also introduce much useless and potentially interfering information. To alleviate the above issues, we propose a simple yet robust tracker that separates temporal information learning from appearance modeling and extracts temporal relations from a set of representative tokens rather than several images (or features). Specifically, we introduce one track token for each frame to collect the target's appearance information in the backbone. Then, we design a mamba-based Temporal Module for track tokens to be aware of context by interacting with other track tokens within a sliding window. This module consists of a mamba layer with autoregressive characteristic and a cross-attention layer with strong global perception ability, ensuring sufficient interaction for track tokens to perceive the appearance changes and movement trends of the target. Finally, track tokens serve as a guidance to adjust the appearance feature for the final prediction in the head. Experiments show our method is effective and achieves competitive performance on multiple benchmarks at a real-time speed.

Jingjing Xie, Yuxin Zhang, Jun Peng, Zhaohong Huang, Liujuan Cao

Despite the efficiency of prompt learning in transferring vision-language models (VLMs) to downstream tasks, existing methods mainly learn the prompts in a coarse-grained manner where the learned prompt vectors are shared across all categories. Consequently, the tailored prompts often fail to discern class-specific visual concepts, thereby hindering the transferred performance for classes that share similar or complex visual attributes. Recent advances mitigate this challenge by leveraging external knowledge from Large Language Models (LLMs) to furnish class descriptions, yet incurring notable inference costs. In this paper, we introduce TextRefiner, a plug-and-play method to refine the text prompts of existing methods by leveraging the internal knowledge of VLMs. Particularly, TextRefiner builds a novel local cache module to encapsulate fine-grained visual concepts derived from local tokens within the image branch. By aggregating and aligning the cached visual descriptions with the original output of the text branch, TextRefiner can efficiently refine and enrich the learned prompts from existing methods without relying on any external expertise. For example, it improves the performance of CoOp from 71.66% to 76.96% on 11 benchmarks, surpassing CoCoOp which introduced instance-wise feature for text prompts. Equipped with TextRefiner, PromptKD achieves state-of-the-art performance while keep inference efficient.

Jianxiang Xie, Yao Wu, Yachao Zhang, Zhongchao Shi, Jianping Fan, Yuan Xie, Yanyun Qu

Source-Free Domain Adaptation (SFDA) aims to transfer a pre-trained source model to the unlabeled target domain without accessing the source data, thereby effectively solving labeled data dependency and domain shift problems. However, the SFDA setting faces a bottleneck due to the absence of supervisory information. To mitigate this problem, Active Learning (AL) is introduced to combine with SFDA, endeavoring to actively label a small set of the most high-quality target points so that models with satisfactory performance can be obtained at an acceptable cost. Nevertheless, several issues remain unresolved, namely when to query new labels during training, what kind of samples deserve labeling to ensure rich information, and where the labels should be distributed to guarantee diversity. Thus we elaborate OmniQuery to omnibearing address the “When, What, and Where” problems about active points querying in source-free domain adaptation for cross-modal 3D semantic segmentation. The method consists of three main components: Query Decider, Point Ranker, and Budget Slicer. The Query Decider determines the optimal timing to query new points by fitting the validation curves during training. The Point Ranker nominates points for annotation by calculating the ambiguity of neighboring points in the feature space. The Budget Slicer allocates the annotation quota, i.e., labeling percentage of the point cloud, to different semantic regions by utilizing the advanced 2D semantic segmentation capabilities of the Segment Anything Model (SAM). Extensive experiments demonstrate the effectiveness of our proposed method, achieving up to 99.64% of fully supervised performance with only 3% of labels, and consistently outperforming comparison methods across various scenarios.

Zeyu Xiao, Zhuoyuan Li, Wei Jia

Transformer-based networks have set new benchmarks in light field super-resolution (SR), but adapting them to capture both global and local spatial-angular correlations efficiently remains challenging. Moreover, many methods fail to account for geometric details like occlusions, leading to performance drops. To tackle these issues, we introduce OHT. This hybrid network leverages occlusion maps through an occlusion-embedded mix layer. It combines the strengths of convolutional networks and Transformers via spatial-angular separable convolution (SASep-Conv) and angular self-attention (ASA). SASep-Conv offers a lightweight alternative to 3D convolution for capturing spatial-angular correlations, while the ASA mechanism applies 3D self-attention across the angular dimension. These designs allow OHT to capture global angular correlations effectively. Extensive experiments on multiple datasets demonstrate OHT's superior performance.