Progressive Relation Learning for Group Activity Recognition | |
Guyue, Hu1,2; Bo, Cui1,2; Yuan, He1,2; Shan, Yu1,2,3 | |
2020 | |
会议日期 | JUN 14-19, 2020 |
会议地点 | ELECTR NETWORK |
DOI | 10.1109/CVPR42600.2020.00106 |
页码 | 977-986 |
英文摘要 | Group activities usually involve spatiotemporal dynamics among many interactive individuals, while only a few participants at several key frames essentially define the activity. Therefore, effectively modeling the group-relevant and suppressing the irrelevant actions (and interactions) are vital for group activity recognition. In this paper, we propose a novel method based on deep reinforcement learning to progressively refine the low-level features and high-level relations of group activities. Firstly, we construct a semantic relation graph (SRG) to explicitly model the relations among persons. Then, two agents adopting policy according to two Markov decision processes are applied to progressively refine the SRG. Specifically, one featured-istilling (FD) agent in the discrete action space refines the low-level spatiotemporal features by distilling the most informative frames. Another relation-gating (RG) agent in continuous action space adjusts the high-level semantic graph to pay more attention to group-relevant relations. The SRG, FD agent, and RG agent are optimized alternately to mutually boost the performance of each other. Extensive experiments on two widely used benchmarks demonstrate the effectiveness and superiority of the proposed approach. |
语种 | 英语 |
内容类型 | 会议论文 |
源URL | [http://ir.ia.ac.cn/handle/173211/44319] |
专题 | 自动化研究所_脑网络组研究中心 |
通讯作者 | Guyue, Hu |
作者单位 | 1.CAS Ctr Excellence Brain Sci & Intelligence Techn, Beijing, Peoples R China 2.Chinese Acad Sci CASIA, Natl Lab Pattern Recognit, Brainnetome Ctr, Beijing, Peoples R China 3.Univ Chinese Acad Sci, Beijing, Peoples R China |
推荐引用方式 GB/T 7714 | Guyue, Hu,Bo, Cui,Yuan, He,et al. Progressive Relation Learning for Group Activity Recognition[C]. 见:. ELECTR NETWORK. JUN 14-19, 2020. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
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。
修改评论