Action Segmentation on HOI4D official (test)
85.5AccuracyMAMBA4D
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| MAMBA4DLearning paradigm=Supervised Learning, Reference=CVPR20252026.02 | 85.5 | 91.3 | — | — | 85.5 | |
| X4D-SceneFormer + Mamba4Denhancement=Mamba4D2026.06 | 85.5 | 91.3 | 92.6 | 91.2 | 85.5 | |
| X4DLearning paradigm=Self-Supervised Learning (End-to-End Fine-Tuning), Reference=AAAI20242026.02 | 84.1 | 91.1 | — | — | 84.8 | |
| X4D-SceneFormer2026.06 | 84.1 | 91.1 | 92.5 | 90.8 | 84.8 | |
| CrossVideoLearning paradigm=Self-Supervised Learning (End-to-End Fine-Tuning), Reference=ICRA20242026.02 | 83.7 | 86 | — | — | 76 | |
| PST-Transformer + MotionPDEBackbone=PST-Transformer2026.06 | 83.21 | 76.4 | 79.39 | 77.36 | 71 | |
| PPTr + MotionPDEBackbone=PPTr2026.06 | 82.42 | 87.24 | 88.52 | 86.62 | 80.1 | |
| PPTr+C2PFrames=1502022.12 | 81.1 | 84 | 85.4 | 82.5 | 74.1 | |
| PPTr+C2PLearning paradigm=Self-Supervised Learning (End-to-End Fine-Tuning), Reference=CVPR20232026.02 | 81.1 | 84 | — | — | 74.1 | |
| PPTr + C2Penhancement=C2P2026.06 | 81.1 | 84 | 85.4 | 82.5 | 74.1 | |
| P4Trans+ATALearning paradigm=Adaptation for Pre-Trained Models, Reference=-2026.02 | 79.9 | 74.7 | — | — | 65 | |
| PPTr + Leafenhancement=Leaf2026.06 | 79.4 | 83.9 | 85 | 81.9 | 73.3 | |
| PST-Transformerimplementation=our own implementation2026.06 | 79.22 | 59.91 | 65.39 | 62.66 | 54.64 | |
| PPTr+VideoMAEFrames=1502022.12 | 78.6 | 80.2 | 81.9 | 78.7 | 69.9 | |
| PPTr+STRLFrames=1502022.12 | 78.4 | 79.1 | 81.8 | 78.6 | 69.7 | |
| PPTrFrames=1502022.12 | 77.4 | 80.1 | 81.7 | 78.5 | 69.5 | |
| PPTrLearning paradigm=Supervised Learning, Reference=ECCV20222026.02 | 77.4 | 80.1 | — | — | 69.5 | |
| PPTr2026.06 | 77.4 | 80.1 | 81.7 | 78.5 | 69.5 | |
| P4Transformer+C2PFrames=1502022.12 | 73.5 | 76.8 | 77.2 | 72.9 | 62.4 | |
| P4Trans+C2PLearning paradigm=Self-Supervised Learning (End-to-End Fine-Tuning), Reference=CVPR20232026.02 | 73.5 | 76.8 | — | — | 62.4 | |
| P4TransformerFrames=1502022.12 | 71.2 | 73.1 | 73.8 | 69.2 | 58.2 | |
| P4TransformerLearning paradigm=Supervised Learning, Reference=CVPR20212026.02 | 71.2 | 73.1 | — | — | 58.2 | |
| P4Transformer2026.06 | 71.2 | 73.1 | 73.8 | 69.2 | 58.2 | |
| Asformer2026.06 | 46.8 | 80.3 | 58.9 | 51.3 | 35 | |
| MS-TCN2026.06 | 44.2 | 74.7 | 55.6 | 47.8 | 31.8 | |
| MS-TCN++2026.06 | 42.2 | 75.8 | 54.7 | 46.5 | 30.3 |