Text-to-3D Generation on SceneBench Scene-level
64.87Imaging ScoreVIST3A
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| VIST3Avariant=Wan + AnySplat2025.10 | 64.87 | 56.96 | 30.18 | 3.67 | 3.86 | 3.4 | |
| VIST3Avariant=Wan + MVDUSt3R2025.10 | 62.08 | 55.67 | 30.26 | 3.72 | 3.97 | 3.47 | |
| VideoRFSplat2025.10 | 58.19 | 51.71 | 29.76 | 3.58 | 3.63 | 3.3 | |
| SplatFlow2025.10 | 48.85 | 53.71 | 29.43 | 3.47 | 3.65 | 3.26 | |
| Director3D2025.10 | 47.79 | 52.81 | 29.31 | 3.36 | 3.67 | 3.2 | |
| Matrix3D-omni2025.10 | 46.65 | 37.62 | 24.04 | 2.66 | 3.29 | 2.8 | |
| Prometheus3D2025.10 | 44.73 | 45.85 | 28.57 | 3.2 | 3.36 | 2.98 |