List your company
ResearchFormerly WizWand

Facial Expression Recognition on AFEW (test)

60.05AccuracyARPGNet
37.01442.994548.97554.9555Nov 12, 2018Jan 14, 2020Mar 18, 2021May 21, 2022Jul 23, 2023Sep 24, 2024Nov 27, 2025
Updated 8mo ago

Evaluation results

MethodLinks
ARPGNetStream type=Multi-stream, Configuration=backbone ablation2025.1160.05
HSE-NNStream type=Single-stream2025.1159.3
ARPGNetStream type=Multi-stream, Backbone=default, Fusion method comparison=controlled backbone2025.1157.7
5CNNsStream type=Multi-stream2025.1157.4
MSCMStream type=Single-stream2025.1156.4
4CNNs + LMEDStream type=Multi-stream2025.1156.1
MulTStream type=Multi-stream, Fusion method comparison=controlled backbone2025.1155.87
Three-level attention with NSTStream type=Single-stream2025.1155.2
TEMMAStream type=Multi-stream, Fusion method comparison=controlled backbone2025.1154.83
Two-stageStream type=Single-stream2025.1154.8
MDANStream type=Single-stream2025.1154.7
AENStream type=Single-stream2025.1154.6
Expression Snippet TransformerStream type=Single-stream2025.1154.3
PACVTStream type=Multi-stream2025.1154
BLSTMStream type=Single-stream2025.1153.9
APViTStream type=Single-stream2025.1153.8
Statistical encodingStream type=Single-stream2025.1153.5
Multi-attentionStream type=Multi-stream, Fusion method comparison=controlled backbone2025.1153
DSANStream type=Single-stream2025.1152.7
PoolingStream type=Single-stream2025.1152.2
RNN + C3DStream type=Multi-stream2025.1152
AM-CNNStream type=Multi-stream2025.1151.4
Graph-TranStream type=Multi-stream2025.1151.2
Former-DFERStream type=Single-stream2025.1150.9
URNNStream type=Single-stream2025.1149
VGG-FACE-LSTM-external-augmentationarchitecture=LSTM, training_data=Union of AFEW and external data, data_augmentation=true2018.1148.6
VGG-FACE-LSTM-external-augmentationBackbone=VGG-FACE-LSTM, Training Data=AFEW + external data, Augmentation=Yes2018.1148.6
VGG-FACE (Proposed Approach)use_synthesized_images=true, backbone=VGG-FACE2018.1148.4
VGG-FACE trained using the proposed approachBackbone=VGG-FACE, Training Method=Facial affect synthesis framework2018.1148.4
VGG-FACE-FERpre-training=FER2013, fine-tuning=AFEW2018.1148.3
VGG-FACE-FERBackbone=VGG-FACE, Pre-training=FER20132018.1148.3
VGG-FACE-externaltraining_data=Union of AFEW and external data2018.1141.4
VGG-FACE-externalBackbone=VGG-FACE, Training Data=AFEW + external data2018.1141.4
the VGG-FACE baselinetraining_data=AFEW training set2018.1137.9
VGG-FACE baselineBackbone=VGG-FACE, Training Data=AFEW training set2018.1137.9