心理科学进展2025，Vol.33Issue(11)：1837-1853,17.DOI:10.3724/SP.J.1042.2025.1837
面向实际应用的微表情分析:从数据采集到智能部署
Micro-expression analysis for practical applications:From data acquisition to intelligent deployment
摘要
Abstract
Micro-expressions,subtle facial movements that occur when individuals attempt to conceal their true emotions,serve as highly efficient non-invasive cues for monitoring psychological states.They hold immense potential for application in fields such as national and public security.However,implementing micro-expression analysis in specific application scenarios poses numerous challenges.In response to these challenges,an approach is taken that initially designs micro-expression elicitation paradigms with high ecological validity based on physiological and behavioral psychology research.This is complemented by the development of an auxiliary coding system based on facial peripheral electromyography and the creation of a large-scale micro-expression database to deeply explore the physiological and behavioral mechanisms of micro-expressions in interactive environments.Furthermore,efforts are focused on addressing the difficulties of constructing micro-expression analysis models in specific application scenarios.This includes designing interference elimination modules to avoid the impact of head movements and lip movement changes on micro-expression spotting.Additionally,self-supervised recognition models are built based on large-scale vertical domain vision models to address the challenge of small sample sizes.Moreover,considering data privacy and transmission constraints,intelligent micro-expression analysis deployment in different scenarios is realized based on asynchronous federated learning.Through the interdisciplinary integration of psychology and computer science,this endeavor aims to construct a micro-expression analysis framework that balances theory and practical application.This not only enhances the accuracy and applicability of micro-expression analysis but also paves the way for its application in more fields.关键词
微表情智能分析/情感计算/高生态效度/实际应用Key words
micro-expression intelligent analysis/affective computing/high ecological validity/practical application分类
社会科学引用本文复制引用
李婧婷,赵琳,东子朝,王甦菁..面向实际应用的微表情分析:从数据采集到智能部署[J].心理科学进展,2025,33(11):1837-1853,17.基金项目
国家自然科学基金项目(62476269,62276252),中国科学院青年创新促进会. （62476269,62276252）
