Action Recognition on EPIC-Kitchens S2 unseen 2019 (test)
59Top-1 Verb AccAssembleNet++ ODF+SDF
Evaluation Results
| Method | Links | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| AssembleNet++ ODF+SDFSK Size=exact2020.01 | 59 | 35.7 | 27.3 | 83.3 | 59 | 44 | — | — | — | — | — | — | — | — | |
| AssembleNet++ ODF+SDFSK Size=10242020.01 | 58.3 | 35.2 | 25.9 | 82.1 | 58.2 | 42.9 | — | — | — | — | — | — | — | — | |
| WeakLargeScale2020.01 | 57.3 | 35.7 | 25.6 | 81.1 | 58.7 | 42.7 | — | — | — | — | — | — | — | — | |
| DEEP-HAL+ODF+SDFSK Size=10242020.01 | 55.3 | 32.6 | 22.3 | 79.1 | 55.4 | 39.2 | — | — | — | — | — | — | — | — | |
| TBNModel Type=Ensemble2019.08 | 54.46 | 30.39 | 20.97 | 81.23 | 55.69 | 39.4 | 32.57 | 21.68 | 10.96 | 27.6 | 25.58 | 13.31 | — | — | |
| TBNModel Type=Single Model2019.08 | 52.69 | 27.86 | 19.06 | 79.93 | 53.78 | 36.54 | 31.44 | 21.48 | 12 | 28.21 | 23.53 | 12.69 | — | — | |
| LFB MaxPooling=Max2020.01 | 50.9 | 31.5 | 21.2 | 77.6 | 57.8 | 39.4 | — | — | — | — | — | — | — | — | |
| TSN w. AudioBackbone=TSN, Modality=w. Audio2019.08 | 46.61 | 22.5 | 13.05 | 78.19 | 48.59 | 29.13 | 28.92 | 15.48 | 6.47 | 21.58 | 16.61 | 7.55 | — | — | |
| [8] (from leaderboard)Source=leaderboard2019.08 | 39.4 | 22.7 | 10.89 | 74.29 | 45.72 | 25.26 | 22.54 | 15.33 | 6.21 | 13.06 | 17.52 | 6.49 | — | — | |
| TSN Fusion2020.01 | 39.4 | 22.7 | 10.9 | 74.3 | 45.7 | 25.3 | — | — | — | — | — | — | — | — | |
| Attention Clusters2019.08 | 32.37 | 11.95 | 5.6 | 69.89 | 31.82 | 15.74 | 17.21 | 3.86 | 1.84 | 11.59 | 7.94 | 2.64 | — | — | |
| Action BanksInput Modality=RGB, Number of frames=8, Number of crops=10, Resolution=full res2020.04 | — | — | — | — | — | — | — | — | — | — | — | — | 21.2 | 39.4 | |
| TRNInput Modality=RGB, Number of frames=8, Number of crops=102020.04 | — | — | — | — | — | — | — | — | — | — | — | — | 15.7 | 30 | |
| TRN-MultiscaleInput Modality=RGB, Number of frames=8, Number of crops=102020.04 | — | — | — | — | — | — | — | — | — | — | — | — | 16.3 | 29.7 | |
| TSMInput Modality=RGB, Number of frames=8, Number of crops=102020.04 | — | — | — | — | — | — | — | — | — | — | — | — | 15 | 28.1 | |
| TSM+W3Input Modality=RGB, Number of frames=8, Number of crops=102020.04 | — | — | — | — | — | — | — | — | — | — | — | — | 17.8 | 32.6 | |
| TSM+W3Input Modality=RGB, Number of frames=8, Number of crops=10, Resolution=full res2020.04 | — | — | — | — | — | — | — | — | — | — | — | — | 18.7 | 33.7 | |
| TSNInput Modality=RGB, Number of frames=8, Number of crops=102020.04 | — | — | — | — | — | — | — | — | — | — | — | — | 12.8 | 26.1 |