Comparison of Simplified SE-ResNet and SE-DenseNet for Micro-Expression Classification

Micro-expressions are rapid and subtle facial movements that can reflect the most real emotional state hidden in the human heart. Classifying different micro-expressions is still challenging because of their short duration and low intensity. This paper proposes new neural network models, Simplified SE-DenseNet-cc and SE-ResNet-cc, incorporating Eulerian video magnification (EVM) to enlarge micro-expression movements. Important features can be selectively enhanced, and unimportant features can be compressed using SE-block. The experimental results show that our proposed methods perform better than most of the algorithms in CASME-II and SMIC.