Moment Query on Ego4D Moment Query (val)
32.48Avg mAPInternVideo + EgoVideo-V + EgoVideo-MQ
Evaluation results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| InternVideo + EgoVideo-V + EgoVideo-MQFeature=InternVideo + EgoVideo-V + EgoVideo-MQ2024.06 | 32.48 | — | — | 51.04 | — | — | — | — | — | — | |
| CausalTADParams (M)=76.982026.04 | 32.19 | — | — | — | — | — | — | — | — | — | |
| InternVideo + EgoVideo-VFeature=InternVideo + EgoVideo-V2024.06 | 31.3 | — | — | 50.21 | — | — | — | — | — | — | |
| InternVideo + EgoVideo-MQFeature=InternVideo + EgoVideo-MQ2024.06 | 31 | — | — | 49.28 | — | — | — | — | — | — | |
| EgoVideo-MQFeature=EgoVideo-MQ2024.06 | 28.53 | — | — | 46.07 | — | — | — | — | — | — | |
| InternVideo + EgoVLPFeature=InternVideo + EgoVLP2024.06 | 27.85 | — | — | 46.98 | — | — | — | — | — | — | |
| ASL2023.12 | 27.85 | — | — | 46.98 | — | — | — | — | — | — | |
| LiquidTADParams (M)=11.51, FLOPs (G)=21.722026.04 | 27.81 | — | — | — | — | — | — | — | — | — | |
| ActionFormerParams (M)=40.47, FLOPs (G)=38.072026.04 | 27.63 | — | — | — | — | — | — | — | — | — | |
| Text-Conditioned ResamplerNumber of frames=1922023.12 | 25.45 | — | — | 43.72 | — | — | — | — | — | — | |
| Text-Conditioned ResamplerNumber of frames=962023.12 | 24.51 | — | — | 42.99 | — | — | — | — | — | — | |
| Intern Video2023.12 | 23.59 | — | — | 41.13 | — | — | — | — | — | — | |
| TriDetParams (M)=14.83, FLOPs (G)=42.112026.04 | 22.83 | — | — | — | — | — | — | — | — | — | |
| ASLFeatures=EgoVLP2025.02 | 22.8 | — | — | — | — | — | — | 29.5 | 23 | 16.1 | |
| ActionFormerFeatures=SlowFast+Omnivore+EgoVLP, Fusion=Proj+Cat2022.11 | 21.4 | — | — | 38.73 | — | — | — | — | — | — | |
| Hier-EgoPackFeatures=EgoVLP2025.02 | 21 | — | — | — | — | — | — | 27 | 21 | 15.2 | |
| ActionFormerFeatures=SlowFast+Omnivore+EgoVLP, Fusion=Cat2022.11 | 20.9 | — | — | 36.84 | — | — | — | — | — | — | |
| ActionFormerFeatures=EgoVLP2025.02 | 20.6 | — | — | — | — | — | — | 26.8 | 20.6 | 14.5 | |
| ActionFormerFeatures=SlowFast+Omnivore, Fusion=Cat2022.11 | 17.17 | — | — | 33.46 | — | — | — | — | — | — | |
| ActionFormerFeatures=SlowFast, Fusion=n/a2022.11 | 13.24 | — | — | 27.8 | — | — | — | — | — | — | |
| VSGNFeatures=EgoVLP2025.02 | 11.4 | — | — | — | — | — | — | 16.6 | 11.5 | 6.57 | |
| VSGNFeatures=SlowFast, Fusion=n/a2022.11 | 6.03 | — | — | 25.16 | — | — | — | — | — | — | |
| Ego4D BaselineFeatures=SlowFast2025.02 | 6.03 | — | — | — | — | — | — | 9.1 | 5.76 | 3.41 | |
| VSGNVis Enc=Frozen+EgoNCE, Vis-text PT=EgoClip2022.06 | 0.1139 | 0.4043 | 0.6567 | 0.3014 | 0.5198 | 0.1906 | 0.2977 | 0.1663 | 0.1145 | 0.0657 | |
| VSGNVis Enc=Frozen, Vis-text PT=EgoClip2022.06 | 0.1069 | 0.4006 | 0.6371 | 0.2959 | 0.4832 | 0.1741 | 0.2633 | 0.159 | 0.1054 | 0.0619 | |
| VSGNVis Enc=Frozen, Vis-text PT=HowTo100M2022.06 | 0.0672 | 0.314 | 0.5261 | 0.2228 | 0.4129 | 0.1341 | 0.2321 | 0.0983 | 0.0672 | 0.0384 | |
| VSGNVis Enc=Frozen, Vis-text PT=CC3M+WebVid-2M2022.06 | 0.0658 | 0.3208 | 0.564 | 0.2346 | 0.4381 | 0.1373 | 0.2377 | 0.0983 | 0.064 | 0.0386 | |
| VSGNVis Enc=SlowFast, Vis-text PT=-2022.06 | 0.0603 | 0.3345 | 0.5843 | 0.2516 | 0.4618 | 0.1536 | 0.2581 | 0.091 | 0.0576 | 0.0341 |