3D Object Captioning on 3D Objects
57.63Sentence-BERT ScoreGPT-4V
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| GPT-4VInput=Single-view Img.2025.12 | 57.63 | 58.72 | 56.89 | |
| Qwen2.5-VL-7BTrainable Params=7.61B, Input=Single-view Img.2025.12 | 52.74 | 54.33 | 52.04 | |
| Lemon-7BTrainable Params=7.63B, Input=3D Point Cloud2025.12 | 52.23 | 53.59 | 50.76 | |
| GreenPLMTrainable Params=3.8B, Input=3D Point Cloud2025.12 | 48.72 | 48.4 | 42.78 | |
| ShapeLLM-13BTrainable Params=13.04B, Input=3D Point Cloud2025.12 | 47.8 | 49.21 | 46.09 | |
| PointLLM-13BTrainable Params=13.01B, Input=3D Point Cloud2025.12 | 47.67 | 48.22 | 40.39 | |
| MiniGPT-3DTrainable Params=2.7B, Input=3D Point Cloud2025.12 | 47.64 | 47.2 | 45.78 | |
| ShapeLLM-7BTrainable Params=7.04B, Input=3D Point Cloud2025.12 | 47.63 | 49.35 | 45.81 | |
| PointLLM-7BTrainable Params=7.01B, Input=3D Point Cloud2025.12 | 47.33 | 47.93 | 40.78 | |
| 3D-LLMInput=Multi-view Img.2025.12 | 42.13 | 42.79 | 32.6 | |
| LLaVA-1.5-13BTrainable Params=13.03B, Input=Single-view Img.2025.12 | 38.89 | 40.54 | 17.2 |