Relational Prototypical Network for Weakly Supervised Temporal Action Localization | |
Huang, Linjiang1,3; Huang, Yan1,3; Ouyang, Wanli4; Wang, Liang1,2,3 | |
2020-02 | |
会议名称 | Thirty-Fourth AAAI Conference on Artificial Intelligence |
会议日期 | 2020-2-7 |
会议地点 | New York, USA |
摘要 | In this paper, we propose a weakly supervised temporal action localization method on untrimmed videos based on prototypical networks. We observe two challenges posed by weakly supervision, namely action-background separation and action relation construction. Unlike the previous method, we propose to achieve action-background separation only by the original videos. To achieve this, a clustering loss is adopted to separate actions from backgrounds and learn intra-compact features, which helps in detecting complete action instances. Besides, a similarity weighting module is devised to further separate actions from backgrounds. To effectively identify actions, we propose to construct relations among actions for prototype learning. A GCN-based prototype embedding module is introduced to generate relational prototypes. Experiments on THUMOS14 and ActivityNet1.2 datasets show that our method outperforms the state-of-the-art methods. |
收录类别 | EI |
资助项目 | Capital Science and Technology Leading Talent Training Project[Z181100006318030] ; National Natural Science Foundation of China[61806194] ; National Natural Science Foundation of China[61420106015] ; National Natural Science Foundation of China[61721004] ; National Natural Science Foundation of China[61633021] ; National Natural Science Foundation of China[61525306] ; National Key Research and Development Program of China[2016YFB1001000] |
语种 | 英语 |
七大方向——子方向分类 | 目标检测、跟踪与识别 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/39128 |
专题 | 智能感知与计算研究中心 |
通讯作者 | Wang, Liang |
作者单位 | 1.Center for Research on Intelligent Perception and Computing, National Laboratory of Pattern Recognition 2.Center for Excellence in Brain Science and Intelligence Technology, Institute of Automation, Chinese Academy of Sciences 3.University of Chinese Academy of Sciences 4.University of Sydney |
通讯作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Huang, Linjiang,Huang, Yan,Ouyang, Wanli,et al. Relational Prototypical Network for Weakly Supervised Temporal Action Localization[C],2020. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
AAAI-HuangL.1235.pdf(859KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
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