期刊:2019 IEEE 1st Global Conference on Life Sciences and Technologies (LifeTech)日期:2019-03-01卷期号:58: 85-86被引量:1
标识
DOI:10.1109/lifetech.2019.8884040
摘要
In a facial expression recognition model that we previously proposed, additional learning of facial expression patterns is performed based on a feature space generated by initial learning; therefore, properly setting initial learning data is necessary. This paper defines an index for measuring dispersion of facial expression patterns and analyzes the relationship between the dispersion and the accuracy of facial expression recognition using classifiers generated through additional learning.