基于光流和卷积视觉Transformer的轻量级微表情识别

徐恺蔚,KHIZER BIN TALIBHafiz,曹衍龙,许源平,许志杰,宋景春

Lightweight micro-expression recognition based on optical flow and convolutional vision Transformer

Kaiwei XU,Hafiz KHIZER BIN TALIB,Yanlong CAO,Yuanping XU,Zhijie XU,Jingchun SONG

表 3 模型复杂度及性能对比

Tab.3 Comparison of model complexity and performance

模型 Np/M FLOPs/G Full UF1 ViT-16/B[21] 86.6 17.6 0.8512 ResNet-18[12] 11.7 1.8 0.8696 EDSMISEViTNet[27] 3.9 — 0.7587 MiER-CvT 7.5 0.1 0.9102