PSYCH OpenIR  > 中国科学院心理健康重点实验室
细粒度分层时空特征描述符的微表情识别方法
Alternative TitleFine-Grained Hierarchical Spatiotemporal Descriptors for Micro-Expression Recognition
张力为1; 王甦菁2; 段先华1
First Author张力为
Contribution Rank2
Abstract

由于微表情持续时间小于0.5 s、非自愿性和低强度等特点,微表情识别仍然是具有挑战性的任务。对分层时空特征描述符进行改进,提出一种新的细粒度分层时空特征的微表情识别方法。首先提取微表情视频片段中的各层次时空特征,利用投影矩阵建立时空特征和微表情之间的联系,进而选择对识别任务有贡献的区域。然后统计具有整体最大贡献度的层次,将该层次下选中的区域块和前一层选中的区域块进行交集操作,达到去除分层时空特征的空间冗余性和提升微表情特征区分度的目的。在CASMEⅡ上的实验表明提出的方法能够细粒度化微表情发生区域,获得了更好的识别结果。

Other Abstract

Micro-expression recognition is still a challenging task due to its short duration of less than 0.5 s, involuntariness, and low intensity. The hierarchical spatio-temporal feature descriptor is improved, and a new finegrained hierarchical spatio-temporal feature micro-expression recognition method is proposed. Firstly, the spatiotemporal features of each level in the micro-expression video clip are extracted, and the projection matrix is used to establish the relationship between the spatio-temporal features and the micro-expressions, and then the regions that contribute to the recognition task are selected. Then count the layer with the overall maximum contribution, and perform the intersection operation between the selected blocks at the lower layer and the selected blocks in the previous layer to achieve the purpose of removing the spatial redundancy of the hierarchical spatiotemporal features and improving the discrimination of the micro-expression features. Experiments on CASMEⅡ show that the proposed method can fine-grain the micro-expression area and obtain better recognition results.

Keyword微表情识别 分层时空特征 细粒度
2021
Language中文
DOI10.3778/j.issn.1002-8331.2003-0431
Source Publication计算机工程与应用
ISSN1002-8331
Pages9
Subtype综述
Indexed ByCSCD
Project Intro.

国家自然科学基金(U19B2032,61772511)资助

CSCD IDCSCD:7029626
Citation statistics
Document Type期刊论文
Identifierhttp://ir.psych.ac.cn/handle/311026/38965
Collection中国科学院心理健康重点实验室
Affiliation1.江苏科技大学计算机学院
2.中国科学院心理研究所行为科学重点实验室
Recommended Citation
GB/T 7714
张力为,王甦菁,段先华. 细粒度分层时空特征描述符的微表情识别方法[J]. 计算机工程与应用,2021:9.
APA 张力为,王甦菁,&段先华.(2021).细粒度分层时空特征描述符的微表情识别方法.计算机工程与应用,9.
MLA 张力为,et al."细粒度分层时空特征描述符的微表情识别方法".计算机工程与应用 (2021):9.
Files in This Item:
File Name/Size DocType Version Access License
细粒度分层时空特征描述符的微表情识别方法(905KB)期刊论文出版稿限制开放CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[张力为]'s Articles
[王甦菁]'s Articles
[段先华]'s Articles
Baidu academic
Similar articles in Baidu academic
[张力为]'s Articles
[王甦菁]'s Articles
[段先华]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[张力为]'s Articles
[王甦菁]'s Articles
[段先华]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.