返回
Recognizing spontaneous micro-expression from eye region
DOI:10.1016/j.neucom.2016.03.090.png)
摘要
En 中文
Micro-expression is a kind of spontaneous facial expression, which is with short duration and low intensity. Because of its involuntary feature, it is helpful to reveal one's true emotion when someone tries to conceal. Therefore, it has attracted a great of attentions from the field of affective computing. Previous methods focus on recognizing micro-expression on the whole face. In fact, it is worthy to note that micro-expression often appears in the eye area. In this paper, we present a framework to recognize micro-expressions within the eye region, namely eyeME. Specifically, the LBP-TOP feature is extracted from the eye region, and multiple classifiers are trained to recognize the expressions. We test the proposed framework on the widely used CASME2 database. The experimental results showed that the proposed eyeME framework performs better than the methods using the whole face and mouth region when identifying happy and disgust expressions. It confirmed that the information on eye region is critical to the recognition of these kinds of micro-expressions. (C) 2016 Elsevier B.V. All rights reserved.
Keyword:
Micro-expression recognition
Eye
Local binary patterns
Spontaneous expression
AI总结
对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
6.5
论文数:
2.5W
被引数:
6.5W
机构
引用论文
Bilateral stress fractures of the femoral neck associated with abnormal anatomy—a case report股骨颈双侧应力性骨折与异常解剖结构相关——病例报告
Injury
IF0

