Multimodal Classification on CREMA-D (test)
80.21Multi AccuracyPDMP
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| PDMPBackbone=ResNet-18, Modality=Multimodal (Audio + Visual), Number of frames=1, Visual input resolution=224x224, Audio input size=257x299, Batch size=64, Epochs=1002026.04 | 80.21 | — | — | 80.34 | |
| TCMaxFusion Head=Concat2026.02 | 77.6 | 61.7 | 59 | — | |
| TCMaxFusion Head=Share Head2026.02 | 77.5 | 61.6 | 59.1 | — | |
| OGM-GE + OPM2026.02 | 75.7 | 59.1 | 57.5 | — | |
| Unimodal Ensemble2026.02 | 75 | 61.8 | 61.6 | — | |
| D&RBackbone=ResNet-18, Modality=Multimodal (Audio + Visual), Number of frames=1, Visual input resolution=224x224, Audio input size=257x299, Batch size=64, Epochs=1002026.04 | 73.52 | — | — | 73.96 | |
| MLABackbone=ResNet-18, Modality=Multimodal (Audio + Visual), Number of frames=1, Visual input resolution=224x224, Audio input size=257x299, Batch size=64, Epochs=1002026.04 | 73.21 | — | — | 73.77 | |
| MLA2026.02 | 72.5 | 60.9 | 61.8 | — | |
| QMF2026.02 | 72.2 | 63.1 | 60.8 | — | |
| OPM2026.02 | 71.4 | 59.9 | 55.7 | — | |
| MMPareto2026.02 | 70.3 | 64.4 | 65.8 | — | |
| MMParetoBackbone=ResNet-18, Modality=Multimodal (Audio + Visual), Number of frames=1, Visual input resolution=224x224, Audio input size=257x299, Batch size=64, Epochs=1002026.04 | 70.19 | — | — | 70.82 | |
| OGM-GE2026.02 | 69.3 | 51.5 | 35.5 | — | |
| Grad-BlendingBackbone=ResNet-18, Modality=Multimodal (Audio + Visual), Number of frames=1, Visual input resolution=224x224, Audio input size=257x299, Batch size=64, Epochs=1002026.04 | 68.81 | — | — | 69.34 | |
| AGMBackbone=ResNet-18, Modality=Multimodal (Audio + Visual), Number of frames=1, Visual input resolution=224x224, Audio input size=257x299, Batch size=64, Epochs=1002026.04 | 67.21 | — | — | 68.04 | |
| Concat2026.02 | 66.2 | 59.3 | 27.1 | — | |
| AGM2026.02 | 66.2 | 59.3 | 27.2 | — | |
| PMRBackbone=ResNet-18, Modality=Multimodal (Audio + Visual), Number of frames=1, Visual input resolution=224x224, Audio input size=257x299, Batch size=64, Epochs=1002026.04 | 65.12 | — | — | 65.91 | |
| OGM-GEBackbone=ResNet-18, Modality=Multimodal (Audio + Visual), Number of frames=1, Visual input resolution=224x224, Audio input size=257x299, Batch size=64, Epochs=1002026.04 | 64.34 | — | — | 64.93 | |
| Share Head2026.02 | 63.6 | 60.4 | 23.6 | — | |
| FiLM2026.02 | 63.1 | — | — | — | |
| Visual-onlyBackbone=ResNet-18, Modality=Visual, Number of frames=1, Visual input resolution=224x224, Batch size=64, Epochs=1002026.04 | 62.17 | — | — | 62.78 | |
| BiGated2026.02 | 61.6 | — | — | — | |
| ConcatenationBackbone=ResNet-18, Modality=Multimodal (Audio + Visual), Number of frames=1, Visual input resolution=224x224, Audio input size=257x299, Batch size=64, Epochs=1002026.04 | 58.83 | — | — | 59.43 | |
| Audio-onlyBackbone=ResNet-18, Modality=Audio, Audio input size=257x299, Batch size=64, Epochs=1002026.04 | 57.27 | — | — | 57.89 |