Table 2.

Comparison of methods on the AFEW (Acted Facial Expressions in the wild) dataset.

Method	Accuracy
Mode variational LSTM [31]	48.83%
MRAN [14]	49.01%
CNN-BLSTM [33]	49.09%
FAN [18]	51.18%
DenseNet-161 [24]	51.40%
Proposed Method w/o Attention	45.05%
Proposed Method w/o GRU	50.90%
Proposed Method	53.44%