Cap3D
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| 3D Object Captioning | Cap3D subset of 3186 3D objects (test) | CLIP Image-Text Score0.312 | 10 | |
| Text-to-3D Generation | Cap3D 2K (test) | FID31.6 | 2 | |
| 3D Captioning | Cap3D Cap | CIDEr134.1 | 2 | |
| Image-to-3D Generation | Cap3D 2K (test) | FID32.6 | 1 | |
| 3D Question Answering | Cap3D QA | Accuracy39.3 | 1 |
Showing 5 of 5 rows