Action Recognition on Ego-Exo4D Bike Repair
29.03Top-1 AccuracyTIME + VideoMAEv2
Evaluation results
| Method | Links | ||
|---|---|---|---|
| TIME + VideoMAEv22026.05 | 29.03 | 1.33 | |
| TIME + VideoMAEv2Model Strategy=TIME + VideoMAEv2, Evaluation Protocol=Video Probing2026.05 | 29.03 | 1.33 | |
| V-JEPA 2Model Strategy=V-JEPA 2, Evaluation Protocol=Video Probing2026.05 | 28.85 | — | |
| VideoMAEv2Model Strategy=VideoMAEv2, Evaluation Protocol=Video Probing2026.05 | 27.7 | — | |
| TIME + V-JEPA 22026.05 | 26.68 | 2.17 | |
| TIME + V-JEPA 2Model Strategy=TIME + V-JEPA 2, Evaluation Protocol=Video Probing2026.05 | 26.68 | -2.17 | |
| TIME + RVM2026.05 | 23.04 | 0.28 | |
| TIME + RVMModel Strategy=TIME + RVM, Evaluation Protocol=Video Probing2026.05 | 23.04 | 0.28 | |
| RVMModel Strategy=RVM, Evaluation Protocol=Video Probing2026.05 | 22.76 | — | |
| TIME + DINOv3frames=4f2026.05 | 20.69 | 5.07 | |
| TIME + DINOv3Model Strategy=TIME + DINOv3 (4f), Number of frames=4f, Evaluation Protocol=Video Probing2026.05 | 20.69 | 5.07 | |
| TIME + CLIPframes=4f2026.05 | 19.31 | 3.46 | |
| TIME + CLIPModel Strategy=TIME + CLIP (4f), Number of frames=4f, Evaluation Protocol=Video Probing2026.05 | 19.31 | 3.46 | |
| CLIPModel Strategy=CLIP (4f), Number of frames=4f, Evaluation Protocol=Video Probing2026.05 | 15.85 | — | |
| DINOv3Model Strategy=DINOv3 (4f), Number of frames=4f, Evaluation Protocol=Video Probing2026.05 | 15.62 | — | |
| TIMEModel Strategy=TIME (Ours), Evaluation Protocol=Video Probing2026.05 | 14.88 | — |