Class-wise boundary regression by uncertainty in temporal action detection

doi:10.1049/ipr2.12599

CORC > 自动化研究所 > 中国科学院自动化研究所 > 精密感知与控制研究中心 > 精密感知与控制

	Class-wise boundary regression by uncertainty in temporal action detection
	Chen, Yunze 1,2; Chen, Mengjuan 2; Gu, Qingyi 2
刊名	IET IMAGE PROCESSING
	2022-08-04
页码	9
ISSN号	1751-9659
DOI	10.1049/ipr2.12599
通讯作者	Gu, Qingyi(qingyi.gu@ia.ac.cn)
英文摘要	Temporal action detection is a crucial aspect of video understanding. It aims to classify the action as well as locate the start and end boundaries of the action in the untrimmed videos. As deep learning is frequently utilized, the accuracy of annotation is crucial to boundary localization. However, it is observed that some annotation instances are ambiguous and the ambiguity varies between categories. To solve the problem above, a Gaussian model is built to estimate the boundary uncertainty for each instance. Based on instance uncertainty, category uncertainty is applied to describe the uncertainty of each category. By combining instance and category uncertainty, the boundaries of the selected proposals are refined and the ranking of candidate proposals is adjusted. Furthermore, overcorrection is avoided for categories with a high level of uncertainty. With the uncertainty approach, state-of-the-art performance is achieved: 57.5% on THUMOS14 (mAP@0.5) and 35.4% on ActivityNet (mAP@Avg).
资助项目	Scientific Instrument Developing Project of the Chinese Academy of Sciences[YJKYYQ20200045]
WOS研究方向	Computer Science ; Engineering ; Imaging Science & Photographic Technology
语种	英语
出版者	WILEY
WOS记录号	WOS:000836034700001
资助机构	Scientific Instrument Developing Project of the Chinese Academy of Sciences
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/49827]
专题	精密感知与控制研究中心_精密感知与控制
通讯作者	Gu, Qingyi
作者单位	1.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing, Peoples R China 2.Chinese Acad Sci, Ctr Precis Sensing & Control, Inst Automat, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Chen, Yunze,Chen, Mengjuan,Gu, Qingyi. Class-wise boundary regression by uncertainty in temporal action detection[J]. IET IMAGE PROCESSING,2022:9.
APA	Chen, Yunze,Chen, Mengjuan,&Gu, Qingyi.(2022).Class-wise boundary regression by uncertainty in temporal action detection.IET IMAGE PROCESSING,9.
MLA	Chen, Yunze,et al."Class-wise boundary regression by uncertainty in temporal action detection".IET IMAGE PROCESSING (2022):9.