Spatiotemporal Video Question Answering on DAVIS 61
58.75Spatial AccuracyFeature4X
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Feature4XInference Source=Global Novel Feature2025.03 | 58.75 | 58.25 | 61.32 | 3.42 | |
| Feature4XInference Source=Local Novel Feature2025.03 | 54.5 | 54.37 | 56.29 | 2.81 | |
| Feature4XInference Source=Local Novel View2025.03 | 49.75 | 51.13 | 50.31 | 9.78 | |
| Feature4XInference Source=Input Video View2025.03 | 48.5 | 49.84 | 49.06 | 10.02 | |
| Feature4XInference Source=Global Novel View2025.03 | 47 | 49.19 | 47.48 | 10.06 |