Movie Question Answering on MoVQA
42.6Synopsis AccuracyInternVideo2-6B
Evaluation Results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| InternVideo2-6BBackbone=InternVideo2-6B2024.03 | 42.6 | 27.8 | 39.9 | 44.3 | 42.5 | 41.2 | 40.9 | 39.3 | 38.6 | 40.1 | |
| VideoChatBackbone=Eva-g2024.03 | 33.6 | 24.3 | 34.5 | 36.6 | 35.5 | 32 | 35.3 | 32.9 | 33.3 | 34.7 | |
| Mplug-OwlBackbone=CLIP VIT-L2024.03 | 25.1 | 19.9 | 25.3 | 21.9 | 23.5 | 27.5 | 25.2 | 23.5 | 22.1 | 24.8 | |
| Video-ChatGPTBackbone=CLIP VIT-L2024.03 | 23.8 | 20.2 | 22.1 | 22.1 | 21.4 | 24.1 | 23.4 | 22.7 | 22.3 | 22.9 | |
| OtterBackbone=CLIP VIT-L2024.03 | 22.6 | 20.7 | 19.6 | 26.1 | 24.2 | 21.8 | 23.1 | 22.1 | 21.3 | 22.6 |