Multimodal Emotion Recognition on IEMOCAP Unaligned (test)
84.8Happy AccuracyMulT
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| MulTAlignment=Unaligned, CTC alignment=false2019.06 | 84.8 | 81.9 | 77.7 | 74.1 | 73.9 | 70.2 | 62.5 | 59.7 | |
| Full modality (knowledge-transfer network)Available Modalities=Audio, Language, Vision, Alignment=Unaligned2023.12 | 84.8 | 81.9 | 77.7 | 74.1 | 73.9 | 70.2 | 62.5 | 59.7 | |
| knowledge-transfer networkAlignment=Unaligned, Description=Missing-audio scenario with reconstructed audio2023.12 | 84.4 | 81.7 | 77.7 | 74 | 73.8 | 69.6 | 61.9 | 59.5 | |
| Language and vision (knowledge-transfer network)Available Modalities=Language, Vision, Alignment=Unaligned2023.12 | 82.8 | 81 | 76.9 | 72.8 | 72.6 | 68.2 | 60.9 | 58.4 | |
| MCTNAlignment=Unaligned, CTC alignment=true2019.06 | 80.5 | 77.5 | 72 | 71.7 | 64.9 | 65.6 | 49.4 | 49.3 | |
| CTC + MCTNAlignment=Unaligned2023.12 | 80.5 | 77.5 | 72 | 71.7 | 64.9 | 65.6 | 49.4 | 49.3 | |
| Language only (knowledge-transfer network)Available Modalities=Language, Alignment=Unaligned2023.12 | 79.3 | 78.8 | 74.6 | 71.8 | 71.6 | 67.5 | 60.2 | 55.9 | |
| Vision only (knowledge-transfer network)Available Modalities=Vision, Alignment=Unaligned2023.12 | 77.7 | 72.6 | 69.9 | 68.3 | 68.1 | 64.6 | 57.3 | 51.2 | |
| RAVENAlignment=Unaligned, CTC alignment=true2019.06 | 77 | 76.8 | 67.6 | 65.6 | 65 | 64.1 | 62 | 59.5 | |
| CTC + RAVENAlignment=Unaligned2023.12 | 77 | 76.8 | 67.6 | 65.6 | 65 | 64.1 | 62 | 59.5 | |
| EF-LSTMAlignment=Unaligned, CTC alignment=true2019.06 | 76.2 | 75.7 | 70.2 | 70.5 | 72.7 | 67.1 | 58.1 | 57.4 | |
| LF-LSTMAlignment=Unaligned, CTC alignment=false2019.06 | 72.5 | 71.8 | 72.9 | 70.4 | 68.6 | 67.9 | 59.6 | 56.2 |