View-invariant representation and learning of human action

被引:6
作者
Rao, C [1 ]
Shah, M [1 ]
机构
[1] Univ Cent Florida, Sch Elect Engn & Comp Sci, Comp Vis Lab, Orlando, FL 32816 USA
来源
IEEE WORKSHOP ON DETECTION AND RECOGNITION OF EVENTS IN VIDEO, PROCEEDINGS | 2001年
关键词
video understanding; action recognition; view-invariant representation; spatiotemporal curvature; events; activities;
D O I
10.1109/EVENT.2001.938867
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Automatically understanding human actions from video sequences is a very challenging problem. This involves the extraction of relevant visual information front a video sequence, representation of that information in a suitable form, and interpretation of visual information for the purpose of recognition and learning. In this paper, we first present a view-invariant representation of action consisting of dynamic instants and intervals, which is computed using spatiotemporal curvature of a trajectory. This representation is then used by, our system to learn human actions without any, training. The system automatically segments video into individual actions, and computes view invariant representation for each action. The system is able to incrementally learn different actions starting with no model. It is able to discover different instances of the same action Performed by different people, and in different viewpoints. In order to validate our approach, we present results on video clips in which roughly, 50 actions were performed by five different people in different viewpoints. Our system performed impressively by correctly, interpreting most actions.
引用
收藏
页码:55 / 63
页数:9
相关论文
共 50 条
[21]   A View-Invariant Action Recognition Based on Multi-View Space Hidden Markov Models [J].
Ji, Xiaofei ;
Wang, Ce ;
Li, Yibo .
INTERNATIONAL JOURNAL OF HUMANOID ROBOTICS, 2014, 11 (01)
[22]   Online view-invariant human action recognition using rgb-d spatio-temporal matrix [J].
Hsu, Yen-Pin ;
Liu, Chengyin ;
Chen, Tzu-Yang ;
Fu, Li-Chen .
PATTERN RECOGNITION, 2016, 60 :215-226
[23]   View-invariant Feature using Pose Information and Flexible Matching Algorithm for Action Retrieval [J].
Yoshida, Noboru ;
Liu, Jianquan .
2021 ASIA-PACIFIC SIGNAL AND INFORMATION PROCESSING ASSOCIATION ANNUAL SUMMIT AND CONFERENCE (APSIPA ASC), 2021, :1556-1562
[24]   View-invariance in action recognition [J].
Rao, C ;
Shah, M .
2001 IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, VOL 2, PROCEEDINGS, 2001, :316-322
[25]   Cross-View Action Recognition Using View-Invariant Pose Feature Learned from Synthetic Data with Domain Adaptation [J].
Yang, Yu-Huan ;
Liu, An-Sheng ;
Liu, Yu-Hung ;
Yeh, Tso-Hsin ;
Li, Zi-Jun ;
Fu, Li-Chen .
COMPUTER VISION - ACCV 2018, PT II, 2019, 11362 :431-446
[26]   A fast, invariant representation for human action in the visual system [J].
Isik, Leyla ;
Tacchetti, Andrea ;
Poggio, Tomaso .
JOURNAL OF NEUROPHYSIOLOGY, 2018, 119 (02) :631-640
[27]   Multi-view representation learning for multi-view action recognition [J].
Hao, Tong ;
Wu, Dan ;
Wang, Qian ;
Sun, Jin-Sheng .
JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION, 2017, 48 :453-460
[28]   View invariant human action recognition based on factorization and HMMs [J].
Li, Xi ;
Fukui, Kazuhiro .
IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2008, E91D (07) :1848-1854
[29]   Highly Robust Action Retrieval using View-invariant Pose Feature and Simple yet Effective Query Expansion Method [J].
Yoshida, Noboru ;
Liu, Jianquan .
PROCEEDINGS OF 2022 ASIA-PACIFIC SIGNAL AND INFORMATION PROCESSING ASSOCIATION ANNUAL SUMMIT AND CONFERENCE (APSIPA ASC), 2022, :1269-1277
[30]   View-Invariant Center-of-Pressure Metrics Estimation With Monocular RGB Camera [J].
Du, Chen ;
Graham, Sarah ;
Depp, Colin ;
Nguyen, Truong .
IEEE TRANSACTIONS ON MULTIMEDIA, 2023, 25 :7388-7401