Video Question Answering on AGQA v2 (test)
57.8Object-relationIPRM
Evaluation Results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| IPRMframes=322024.11 | 57.8 | — | — | 48 | 75.6 | 62.4 | 50.7 | 20 | 60.4 | 58.6 | 62.8 | |
| VideoDistilltraining=from scratch2024.04 | 56.32 | 55.09 | 55.12 | 43.3 | 51.19 | 55.74 | 49.08 | 10.16 | 55.8 | — | — | |
| GFframes=322024.11 | 55 | — | — | 44.6 | 53.2 | 59.1 | 52.8 | 14.2 | 55.1 | 56.1 | 62.3 | |
| GF(sup)Backbone=FasterRCNN2024.01 | 54.96 | — | — | 44.62 | 53.24 | 59.13 | 52.8 | 14.17 | 55.08 | 56.07 | 54.17 | |
| GF(uns)Backbone=FasterRCNN2024.01 | 54.31 | — | — | 42.9 | 51.97 | 58.08 | 52.02 | 16.38 | 54.59 | 55.66 | 53.52 | |
| GF(uns)Backbone=S3D2024.01 | 52.93 | — | — | 42.78 | 53.03 | 58.31 | 50.86 | 22.08 | 53.33 | 53.06 | 53.61 | |
| MISTframes=322024.11 | 51.7 | — | — | 42.1 | 67.2 | 60.3 | 54.6 | 19.7 | 54.4 | 56.1 | 54.2 | |
| MIST-CLIPfeature=CLIP (ViT-B/32)2022.12 | 51.68 | 67.18 | 68.99 | 42.05 | 67.24 | 60.33 | 54.62 | 19.69 | 54.39 | — | — | |
| MISTBackbone=CLIP2024.01 | 51.68 | — | — | 42.05 | 67.24 | 60.33 | 54.62 | 19.69 | 54.39 | 50.56 | 58.28 | |
| MIST-CLIP2024.04 | 51.68 | 67.18 | 68.99 | 42.05 | 67.24 | 60.33 | 54.62 | 19.69 | 54.39 | — | — | |
| MIST-AIOfeature=All-in-One-Base2022.12 | 51.43 | 54.67 | 55.37 | 41.34 | 53.14 | 53.49 | 47.48 | 20.18 | 50.96 | — | — | |
| Temp[ATP]frames=322024.11 | 51.2 | — | — | 39.8 | 48.3 | 51.8 | 49.6 | 19 | 49.8 | 50.6 | 58.3 | |
| Temp[ATP]feature=CLIP (ViT-B/32)2022.12 | 50.15 | 49.76 | 46.25 | 39.78 | 48.25 | 51.79 | 49.59 | 18.96 | 49.79 | — | — | |
| TempBackbone=CLIP2024.01 | 50.15 | — | — | 39.78 | 48.25 | 51.79 | 49.59 | 18.96 | 49.79 | — | — | |
| Temp[ATP]2024.04 | 50.15 | 49.76 | 46.25 | 39.78 | 48.25 | 51.79 | 49.59 | 18.96 | 49.79 | — | — | |
| AIOfeature=All-in-One-Base2022.12 | 48.34 | 48.99 | 49.66 | 37.53 | 49.61 | 50.81 | 45.36 | 18.97 | 48.59 | — | — | |
| AIOBackbone=ViT2024.01 | 48.34 | — | — | 37.53 | 49.61 | 50.81 | 45.36 | 18.97 | 48.59 | — | — | |
| AIO2024.04 | 48.34 | 48.99 | 49.66 | 37.53 | 49.61 | 50.81 | 45.36 | 18.97 | 48.59 | — | — | |
| All-in-Oneframes=322024.11 | 48.2 | — | — | 37.5 | 49.6 | 50.8 | 45.4 | 19 | 48.6 | — | — | |
| HCRN2022.12 | 40.33 | 49.86 | 49.85 | 33.55 | 49.7 | 50.01 | 43.84 | 5.52 | 42.11 | — | — | |
| HCRNBackbone=C3D2024.01 | 40.33 | — | — | 33.55 | 49.7 | 50.01 | 43.84 | 5.52 | 42.11 | 36.34 | 47.97 | |
| HCRN2024.04 | 40.33 | 49.86 | 49.85 | 33.55 | 49.7 | 50.01 | 43.84 | 5.52 | 42.11 | — | — | |
| HCRNframes=322024.11 | 40.3 | — | — | 33.6 | 49.7 | 50 | 43.8 | 5.5 | 42.1 | — | 48 | |
| PSAC2022.12 | 37.84 | 49.95 | 50 | 33.2 | 49.78 | 49.94 | 45.21 | 4.14 | 40.18 | — | — | |
| PSAC2024.04 | 37.84 | 49.95 | 50 | 33.2 | 49.78 | 49.94 | 45.21 | 4.14 | 40.18 | — | — | |
| HME2022.12 | 37.42 | 49.9 | 49.97 | 33.21 | 49.77 | 49.96 | 47.03 | 5.43 | 39.89 | — | — | |
| HMEBackbone=C3D2024.01 | 37.42 | — | — | 33.21 | 49.77 | 49.96 | 47.03 | 5.43 | 39.89 | 31.01 | 48.91 | |
| HME2024.04 | 37.42 | 49.9 | 49.97 | 33.21 | 49.77 | 49.96 | 47.03 | 5.43 | 39.89 | — | — | |
| Most Likely2022.12 | 9.39 | 50 | 50 | 21.01 | 49.78 | 50 | 24.27 | 5.52 | 10.99 | — | — | |
| Most Likely2024.04 | 9.39 | 50 | 50 | 21.01 | 49.78 | 50 | 24.27 | 5.52 | 10.99 | — | — |