Categorical Emotion Recognition on CREMA-D
85.71UARVAEmo
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| VAEmoSSL=true, Modality=Audio+Visual, #Params(M)=392025.05 | 85.71 | 85.68 | |
| HiCMAE-BSSL=true, Modality=Audio+Visual, #Params(M)=812025.05 | 84.91 | 84.89 | |
| HiCMAE-BModality=Video + Audio2026.01 | 84.91 | 84.89 | |
| HICMAE-SSSL=true, Modality=Audio+Visual, #Params(M)=462025.05 | 84.46 | 84.38 | |
| VQ-MAE-AVSSL=true, Modality=Audio+Visual, #Params(M)=302025.05 | 80.4 | — | |
| V-JEPA for FERModality=Video2026.01 | 79.39 | 78.86 | |
| MultiMAE-DERModality=Video + Audio2026.01 | 79.12 | 79.36 | |
| MAE-DFERSSL=true, Modality=Visual, #Params(M)=852025.05 | 77.33 | 77.38 | |
| MAE-DFERModality=Video2026.01 | 77.33 | 77.38 | |
| SVFAPSSL=true, Modality=Visual, #Params(M)=782025.05 | 77.31 | 77.37 | |
| SVFAP-BModality=Video2026.01 | 77.31 | 77.37 | |
| HiCMAE-BModality=Video2026.01 | 77.25 | 77.21 | |
| WavLM-PlusSSL=true, Modality=Audio, #Params(M)=952025.05 | 73.34 | 73.39 | |
| HuBERTSSL=true, Modality=Audio, #Params(M)=952025.05 | 72.72 | 72.57 | |
| MultiMAE-DERModality=Video2026.01 | — | 77.83 |