Facial Expression Manifold Based on Expression Similarity

A strategy is proposed for facial expression recognition under the graph embedding (GE) frameworkThe neighborhood weighted graph based on the expression similarity is constructed to learn the sub-space. In thesub-space, the data distribute on the manifold based on expression semantic. The proposed sub-space method canovercome the difficulties for facial expression recognition caused by the differences in individuals, lightings, posesThe expressions of the facial images in the data set are exploited in a semi-supervised way. Expression similaritybetween two facial images is measured by the dot product of the expression fuzzy membership function vectorsExperimental results on Cohn-Kanade and the data set of this paper demonstrate the effectiveness of the approach.