Text-to-Panorama Generation on Structured3D
32.48CLIP ScoreGuidedSceneGen
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| GuidedSceneGenConfiguration=full, text + Multi-ControlNet2026.03 | 32.48 | 4.54 | 71 | 3.26 | |
| GuidedSceneGenConfiguration=text + Semantic-ControlNet2026.03 | 32.23 | 4.49 | 70 | 3.41 | |
| GuidedSceneGenConfiguration=text + Depth-ControlNet2026.03 | 31.85 | 4.49 | 70 | 3.33 | |
| MVDiffusion2026.03 | 31.28 | 4.1 | 63 | 3.46 | |
| GuidedSceneGenConfiguration=text only2026.03 | 31.17 | 4.45 | 69 | 3.33 | |
| LayerPano3D2026.03 | 30.95 | 4.38 | 64 | 3.24 | |
| Diffusion3602026.03 | 30.73 | 4.8 | 77 | 3.1 |