Video 3D Perception on ScanNet
20.9F1 Score (IoU=0.25)Spatial Encoder
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Spatial EncoderInput=Video detector2026.03 | 20.9 | 6.2 | |
| SpatialLMInput=Point-cloud detector2026.03 | 13.4 | 2.5 | |
| DetAny3DInput=Image detector2026.03 | 11.9 | 2.4 | |
| SceneScriptInput=Point-cloud detector2026.03 | 10.1 | 4.3 | |
| Qwen3-VL-235BInput=MLLMs2026.03 | 8.7 | 2.8 | |
| 3D-MOODInput=Image detector2026.03 | 8.3 | 2.5 | |
| Qwen3-VL-4BInput=MLLMs2026.03 | 7.7 | 2.3 |