中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 计算技术研究所
  3. 中国科学院计算技术研究所
  4. 中国科学院计算技术研究所期刊论文
  5. 英文
STAU: A SpatioTemporal-Aware Unit for Video Prediction and Beyond

文献类型：期刊论文

作者Chang, Zheng1,2,3,4; Zhang, Xinfeng5; Wang, Shanshe4,6; Ma, Siwei4; Gao, Wen3,7
刊名IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
出版日期2025-09-01
卷号47期号:9页码:7916-7929
关键词Spatiotemporal phenomena Correlation Predictive models Stochastic processes Hands Dynamics Visualization Aggregates Training Reliability Spatiotemporal-aware spatiotemporal correlations attention mechanism video prediction and beyond
ISSN号0162-8828
DOI10.1109/TPAMI.2025.3572735
英文摘要Video prediction aims to predict future frames by modeling the complex spatiotemporal dynamics in videos. However, most existing methods only model the temporal information and the spatial information for videos in an independent manner but have not fully explored the correlations between both terms. In this paper, we propose a SpatioTemporal-Aware Unit (STAU) for video prediction and beyond by exploring the significant spatiotemporal correlations in videos. On the one hand, the motion-aware attention weights are learned from the spatial states to help aggregate the temporal states in the temporal domain. On the other hand, the appearance-aware attention weights are learned from the temporal states to help aggregate the spatial states in the spatial domain. In this way, the temporal information and the spatial information can be greatly aware of each other in both domains, during which, the spatiotemporal receptive field can also be greatly broadened for more reliable spatiotemporal modeling. Experiments are not only conducted on video prediction tasks (deterministic and stochastic), but also another task beyond video prediction, the early action recognition task. Experimental results show that the proposed STAU can achieve satisfactory performance on all tasks compared with other methods.
资助项目Funds for International Cooperation and Exchange of the National Natural Science Foundation of China[62461160310] ; Key Program of the National Natural Science Foundation of China[62431011] ; National Natural Science Foundation of China (NSFC)[62025101] ; PCL-CMCC Foundation for science and innovation[2024ZY1C0040]
WOS研究方向Computer Science ; Engineering
语种英语
WOS记录号WOS:001547756300027
出版者IEEE COMPUTER SOC
源URL[http://119.78.100.204/handle/2XEOYT63/41752]  
专题中国科学院计算技术研究所期刊论文_英文
通讯作者Zhang, Xinfeng
作者单位1.Huawei Technol, Shenzhen 518129, Peoples R China
2.Chinese Acad Sci, Inst Comp Technol, Beijing 100190, Peoples R China
3.Univ Chinese Acad Sci, Beijing 100049, Peoples R China
4.Peking Univ, Natl Engn Res Ctr Visual Technol, Beijing 100871, Peoples R China
5.Univ Chinese Acad Sci, Sch Comp Sci & Technol, Beijing 100049, Peoples R China
6.Peking Univ, Informat Technol R&D Innovat Ctr, Shaoxing 312000, Peoples R China
7.Natl Engn Res Ctr Visual Technol, Beijing 100871, Peoples R China
推荐引用方式
GB/T 7714		 Chang, Zheng,Zhang, Xinfeng,Wang, Shanshe,et al. STAU: A SpatioTemporal-Aware Unit for Video Prediction and Beyond[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,2025,47(9):7916-7929.
APA Chang, Zheng,Zhang, Xinfeng,Wang, Shanshe,Ma, Siwei,&Gao, Wen.(2025).STAU: A SpatioTemporal-Aware Unit for Video Prediction and Beyond.IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,47(9),7916-7929.
MLA Chang, Zheng,et al."STAU: A SpatioTemporal-Aware Unit for Video Prediction and Beyond".IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 47.9(2025):7916-7929.

入库方式： OAI收割

来源：计算技术研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。