Learning multi-temporal-scale deep information for action recognition

doi:10.1007/s10489-018-1347-3

CORC > 光电技术研究所 > 中国科学院光电技术研究所 > 光电探测技术研究室（三室）

	Learning multi-temporal-scale deep information for action recognition
	Yao, Guangle 1,2,3; Lei, Tao 1; Zhong, Jiandan 1,2,3; Jiang, Ping 1
刊名	Applied Intelligence
	2019-06-15
卷号	49 期号:6 页码:2017-2029
ISSN号	0924-669X
DOI	10.1007/s10489-018-1347-3
文献子类	期刊论文
英文摘要	Action recognition in video is widely applied in video indexing, intelligent surveillance, multimedia understanding, and other fields. A typical human action contains the spatiotemporal information from various scales. Learning and fusing the multi-temporal-scale information make action recognition more reliable in terms of recognition accuracy. To demonstrate this argument, in this paper, we use Res3D, a 3D Convolution Neural Network (CNN) architecture, to extract information in multiple temporal scales. And in each temporal scale, we transfer the knowledge learned from RGB to 3-channel optical flow (OF) and learn information from RGB and OF fields. We also propose Parallel Pair Discriminant Correlation Analysis (PPDCA) to fuse the multi-temporal-scale information into action representation with a lower dimension. Experimental results show that compared with single-temporal-scale method, the proposed multi-temporal-scale method gains higher recognition accuracy, and spends more time on feature extraction, but less time on classification due to the representation with lower dimension. Moreover, the proposed method achieves recognition performance comparable to that of the state-of-the-art methods. The source code and 3D filter animations are available online: https://github.com/JerryYaoGl/multi-temporal-scale. © 2018, Springer Science+Business Media, LLC, part of Springer Nature.
语种	英语
出版者	Springer New York LLC
内容类型	期刊论文
源URL	[http://ir.ioe.ac.cn/handle/181551/9659]
专题	光电技术研究所_光电探测技术研究室（三室）
作者单位	1.Institute of Optics and Electronics, Chinese Academy of Sciences, Chengdu, China; 2.University of Chinese Academy of Sciences, Beijing, China 3.University of Electronic Science and Technology of China, Chengdu, China;
推荐引用方式 GB/T 7714	Yao, Guangle,Lei, Tao,Zhong, Jiandan,et al. Learning multi-temporal-scale deep information for action recognition[J]. Applied Intelligence,2019,49(6):2017-2029.
APA	Yao, Guangle,Lei, Tao,Zhong, Jiandan,&Jiang, Ping.(2019).Learning multi-temporal-scale deep information for action recognition.Applied Intelligence,49(6),2017-2029.
MLA	Yao, Guangle,et al."Learning multi-temporal-scale deep information for action recognition".Applied Intelligence 49.6(2019):2017-2029.