3D Video Detection on ScanNet Detection v2 (test)
55.3Chair AP3D-RFT-4B
Evaluation Results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 3D-RFT-4BSetting=4-Frame Setting2026.03 | 55.3 | 18.7 | 48.2 | 39.1 | 49.8 | 77.1 | 50 | 86.1 | 54.2 | 38.2 | 43.7 | |
| VG LLM-8BSetting=4-Frame Setting2026.03 | 54 | 17.1 | 46.5 | 39.8 | 47 | 74.1 | 42.1 | 82.5 | 43.4 | 39.6 | 41.2 | |
| VG LLM-4BSetting=4-Frame Setting2026.03 | 49.7 | 13.1 | 41.3 | 39.2 | 44.6 | 71.2 | 33.5 | 83.4 | 41.7 | 35.7 | 38.2 | |
| 3D-RFT-4BSetting=6-Frame Setting2026.03 | 49.2 | 18.5 | 44.9 | 34.6 | 48.3 | 74.1 | 51.4 | 81.1 | 53.4 | 35.7 | 41.7 | |
| VG LLM-8BSetting=6-Frame Setting2026.03 | 48.7 | 17.9 | 44.8 | 38.5 | 46.4 | 75.8 | 40.4 | 83.2 | 43.5 | 38.7 | 40.8 | |
| VG LLM-4BSetting=6-Frame Setting2026.03 | 41.6 | 12.4 | 39.8 | 33.1 | 45 | 70.2 | 33.8 | 80.6 | 39.7 | 34 | 36.4 | |
| Qwen2.5-VL-7BSetting=4-Frame Setting2026.03 | 41.2 | 11.6 | 36.5 | 30.2 | 41.1 | 68.2 | 36.6 | 68.7 | 34.6 | 31 | 32.5 | |
| Qwen2.5-VL-3BSetting=4-Frame Setting2026.03 | 37.7 | 10.2 | 35 | 23.1 | 39 | 64.8 | 32.4 | 68.8 | 32.6 | 27.9 | 30 | |
| Qwen2.5-VL-7BSetting=6-Frame Setting2026.03 | 36.1 | 10.6 | 32.7 | 25 | 40.7 | 64.6 | 38.4 | 68.6 | 31.8 | 28 | 29.6 | |
| Qwen2.5-VL-3BSetting=6-Frame Setting2026.03 | 32.8 | 7.8 | 31.3 | 20.9 | 32.2 | 58.8 | 36.5 | 66.1 | 27.8 | 24.1 | 25.7 |