Text-conditioned 3D asset generation on G-Objaverse (evaluation)
14.63ISMVD-v2.1-3D
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| MVD-v2.1-3DBase Multi-view Diffusion Model=MVDream 2.1, Adapter=LGAA2025.09 | 14.63 | 36.38 | 41.47 | 33.7 | |
| DV-3DBase Multi-view Diffusion Model=Dreamview, Adapter=LGAA2025.09 | 13.91 | 33.75 | 42.63 | 34.36 | |
| DiffSplat2025.09 | 13.75 | — | — | 31.62 | |
| LGM2025.09 | 13.53 | 36 | — | 32.17 | |
| TRELLISPBR Material Generation Pipeline=TexGaussian2025.09 | 13.4 | 27.08 | 44.05 | 31.25 | |
| MVD-v1.5-3DBase Multi-view Diffusion Model=MVDream 1.5, Adapter=LGAA2025.09 | 12.88 | 40.76 | 53.62 | 32.84 | |
| LaRa2025.09 | 12.77 | 38.8 | — | 31.41 | |
| 3DTopia-XL2025.09 | 9.42 | 64.96 | 47.19 | 29.26 |