3D Visual Grounding on ScanRefer Multiple (val)
52Accuracy @ IoU 0.25MCLN
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MCLNSupervision Level=Fully-Supervised, Venue=ECCV’24, Agent=—2025.12 | 52 | 40.8 | |
| G3-LQSupervision Level=Fully-Supervised, Venue=CVPR’24, Agent=—2025.12 | 50.2 | 39.7 | |
| EDASupervision Level=Fully-Supervised, Venue=CVPR’23, Agent=—2025.12 | 49.1 | 37.6 | |
| BUTD-DETRSupervision Level=Fully-Supervised, Venue=ECCV’22, Agent=—2025.12 | 46.6 | 35.1 | |
| 3D-VisTASupervision Level=Fully-Supervised, Venue=ICCV’23, Agent=—2025.12 | 43.7 | 39.1 | |
| ConcreteNetSupervision Level=Fully-Supervised, Venue=ECCV’24, Agent=—2025.12 | 42.4 | 38.4 | |
| 3DVG-TSupervision Level=Fully-Supervised, Venue=ICCV’21, Agent=—2025.12 | 38.4 | 28.7 | |
| SeeGroundSupervision Level=Zero-Shot, Venue=CVPR’25, Agent=Qwen2-VL-72B2025.12 | 34 | 30 | |
| VoGSupervision Level=Zero-Shot, Venue=Ours, Agent=Qwen2-VL-72B2025.12 | 34 | 30.4 | |
| ScanReferSupervision Level=Fully-Supervised, Venue=ECCV’20, Agent=—2025.12 | 32.1 | 21.3 | |
| InstanceReferSupervision Level=Fully-Supervised, Venue=ICCV’21, Agent=—2025.12 | 31.3 | 24.8 | |
| ZSVG3DSupervision Level=Zero-Shot, Venue=CVPR’24, Agent=GPT-4 Turbo2025.12 | 27.7 | 24.6 | |
| VoGSupervision Level=Zero-Shot, Venue=Ours, Agent=Qwen2-VL-2B2025.12 | 25.3 | 21.9 | |
| SeeGroundSupervision Level=Zero-Shot, Venue=CVPR’25, Agent=Qwen2-VL-2B2025.12 | 20.7 | 18.4 | |
| OpenSceneSupervision Level=Zero-Shot, Venue=CVPR’23, Agent=CLIP2025.12 | 11.1 | 4.4 |