3D Scene Generation on KITTI-360
22.9FIDScenDi
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| ScenDiSetting=3D Gen, Conditions=3D Bbox+Road Map, Backbone=WAN2.1-1.3B [63]2026.01 | 22.9 | 0.016 | 262.6 | 0.06 | 0.23 | — | — | — | — | |
| UrbanGenSetting=3D Gen, Conditions=3D Bbox+Semantic Voxel, Backbone=3D GAN2026.01 | 33 | 0.017 | 300.1 | 0.21 | 0.25 | — | — | — | — | |
| ScenDiSetting=3D Gen, Conditions=3D Bbox+Road Map, Backbone=SVD [5]2026.01 | 36.9 | 0.026 | 400.3 | 0.23 | 0.13 | — | — | — | — | |
| PrITTIVariant=DiT-L2025.06 | 71.028 | — | — | — | — | 80.7 | 54.5 | 4.21 | 1.78 | |
| PrITTIVariant=DiT-XL2025.06 | 71.735 | — | — | — | — | 81.2 | 61.6 | 3.964 | 2.11 | |
| PrITTIVariant=DiT-B2025.06 | 73.952 | — | — | — | — | 71.2 | 49.1 | 3.856 | 0.58 | |
| CC3DSetting=3D Gen, Conditions=BEV Layout, Backbone=3D GAN2026.01 | 90.8 | 0.091 | 706.1 | 1.82 | 1.76 | — | — | — | — | |
| XCubeVariant=Level 22025.06 | 94.822 | — | — | — | — | 48.2 | 23 | 3.48 | 3.5 | |
| DiscoSceneSetting=3D Gen, Conditions=3D Bbox, Backbone=3D GAN2026.01 | 135.3 | 0.093 | 2,025.9 | 5.45 | 2.07 | — | — | — | — | |
| XCubeVariant=Level 12025.06 | 140.455 | — | — | — | — | 24.3 | 0.8 | 2.397 | 1.41 | |
| PDDVariant=Level 32025.06 | 159.878 | — | — | — | — | 15.6 | 2.1 | 2.711 | 12.84 | |
| PDDVariant=Level 22025.06 | 166.086 | — | — | — | — | 18.2 | 0 | 1.896 | 1.89 | |
| SemCityVariant=1M2025.06 | 177.623 | — | — | — | — | 13.3 | 6.3 | 3.454 | 2.35 | |
| PDDVariant=Level 12025.06 | 302.256 | — | — | — | — | 0 | 0 | 1.558 | 0.94 |