Facial Expression Recognition on AffWild2 (test)
72.48AccuracyGReFEL
Evaluation Results
| Method | Links | |
|---|---|---|
| GReFELVenue=Current Paper2024.10 | 72.48 | |
| POSTER++Venue='232024.10 | 69.18 | |
| Tr.FERVenue=ICCV'212024.10 | 68.92 | |
| POSTERVenue=IC-W'222024.10 | 67.74 | |
| Deviation Learning NetworkStream type=Single-stream, Backbone=Inception-Resnet2025.11 | 67.7 | |
| LA-NetVenue=ICCV'232024.10 | 66.76 | |
| F2ExpVenue=CVPR'222024.10 | 66.34 | |
| L.OFERVenue=ICCV'232024.10 | 66.02 | |
| DANVenue=Bioinf.'232024.10 | 65.82 | |
| DMUEVenue=CVPR'212024.10 | 63.64 | |
| EACVenue=ECCV'222024.10 | 63.54 | |
| ARPGNetStream type=Multi-stream, Backbone=Inception-Resnet, Pre-training=FaceNet, DFEW, AffectNet2025.11 | 62.8 | |
| RULVenue=NIPS'212024.10 | 62.37 | |
| Eff.FaceVenue=AAAI'212024.10 | 62.21 | |
| SCNVenue=CVPR'202024.10 | 60.55 | |
| RANVenue=TIP'202024.10 | 59.81 | |
| ARPGNetStream type=Multi-stream, Backbone=default2025.11 | 54.7 | |
| MulTStream type=Multi-stream, Fusion method=standard2025.11 | 53.6 | |
| TEMMAStream type=Multi-stream, Fusion method=standard2025.11 | 53.1 | |
| HSE-NNStream type=Single-stream, Pre-training=AffectNet2025.11 | 52.1 | |
| Two-stageStream type=Single-stream2025.11 | 51.7 | |
| PACVTStream type=Multi-stream2025.11 | 51.7 | |
| R(2+1)DStream type=Single-stream2025.11 | 51.5 | |
| MDANStream type=Single-stream2025.11 | 51 | |
| Multi-attentionStream type=Multi-stream, Fusion method=standard2025.11 | 50.8 | |
| APViTStream type=Single-stream2025.11 | 49.4 | |
| Transformer EncoderStream type=Single-stream, Backbone=IResNet1002025.11 | 47.7 | |
| SCAN + CCIStream type=Single-stream2025.11 | 46.5 | |
| ResNet (EmotionNet)Stream type=Single-stream, Backbone=ResNet50, Pre-training=EmotionNet2025.11 | 46.2 | |
| AM-CNNStream type=Multi-stream2025.11 | 45.9 | |
| Former-DFERStream type=Single-stream2025.11 | 44.9 | |
| CBAM + BLSTMStream type=Single-stream, Backbone=ResNet1012025.11 | 43.4 | |
| MobileNetStream type=Single-stream2025.11 | 36.6 |