Large vision--language models (VLMs) are increasingly applied to long-video question answering, yet inference is often bottlenecked by the number of input frames and resulting visual tokens. Naive sparse sampling can miss decisive moments, while purely relevance-driven selection frequently collapses onto near-duplicate frames and sacrifices coverage of temporally distant evidence. We propose a question-adaptive greedy frame selection method that jointly optimizes query relevance and semantic representativeness under a fixed frame budget. Our approach constructs a 1~FPS candidate pool (capped at 1000) with exact timestamp alignment, embeds candidates in two complementary spaces (SigLIP for question relevance and DINOv2 for semantic similarity), and selects frames by greedily maximizing a weighted sum of a modular relevance term and a facility-location coverage term. This objective is normalized, monotone, and submodular, yielding a standard (1-1/e) greedy approximation guarantee. To account for question-dependent trade-offs between relevance and coverage, we introduce four preset strategies and a lightweight text-only question-type classifier that routes each query to its best-performing preset. Experiments on MLVU show consistent accuracy gains over uniform sampling and a strong recent baseline across frame budgets, with the largest improvements under tight budgets.
EchoPilot: Training-Free Ultrasound Video Segmentation via Scale-Space Semantic Prompting and Reliability-Gated Memory
The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));Third Affiliated Hospital of Sun Yat-Sen University(中山大学第三附属医院);Hong Kong Metropolitan University(香港 Metropolitan 大学)
Greedy Gradient Ensemble for Robust Visual Question Answering
Key Lab of Intell. Info. Process., Inst. of Comput. Tech., CAS(中国科学院计算技术研究所智能信息处理重点实验室);University of Chinese Academy of Sciences(中国科学院大学);Kingsoft Cloud(金山云);Peng Cheng Laboratory(鹏城实验室);Huawei Technologies(华为技术有限公司)
DATE: Dynamic Absolute Time Enhancement for Long Video Understanding
Beihang University(北京航空航天大学);Dcar, ByteDance(字节跳动Dcar部门);Qfin Holdings,Inc(Qfin控股公司);MAIS, Institute of Automation, Chinese Academy of Sciences(自动化研究所,中国科学院)
Efficient universal shuffle attack for visual object tracking
Ji Hua Laboratory(季华实验室);Shanghai Engineering Research Center of AI and Robotics(上海人工智能与机器人工程研究中心);Engineering Research Center of AI and Robotics, Ministry of Education, China(中国教育部人工智能与机器人工程研究中心)
GIFT: Global Irreplaceability Frame Targeting for Efficient Video Understanding
Institute of Science and Technology for Brain-inspired Intelligence, Fudan University(复旦大学类脑智能科学与技术研究院);State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University(北京大学计算机学院多媒体信息处理国家重点实验室);Zhejiang University(浙江大学);Alibaba Group Holding Limited(阿里巴巴集团控股有限公司);Future Living Lab of Alibaba(阿里巴巴未来生活实验室)
A Data-driven Approach for Human Pose Tracking Based on Spatio-temporal Pictorial Structure
Indian Statistical Institute(印度统计学院)
在本文中,我们提出了一个数据驱动的方法,在视频数据中的人体姿态跟踪。我们制定的人体姿态跟踪问题作为一个离散的优化问题的基础上的时空图像结构模型和解决这个问题的贪婪框架非常有效。我们提出的模型来跟踪人体的姿势相结合的人从单一的图像和传统的目标跟踪视频中的姿态估计。我们的姿态跟踪目标函数由以下条款组成:一个帧内的一部分的外观,从前一帧到当前帧的时间位移的一部分,和一个部分的空间依赖性与它的父亲在图形结构中的相似性。在基准数据集(VideoPose2,Poses in the Wild and Outdoor Pose)以及我们新构建的ICDPose数据集上的实验评估表明了我们所提出的方法的实用性。
Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所);School of Cyber Security, University of Chinese Academy of Sciences(中国科学院大学网络安全学院);School of Mathematical Sciences, Fudan University(复旦大学数学学院);School of Cyber Science and Technology, Sun Yat-sen University(中山大学网络科学与技术学院)