Image Relighting On Stanford Orb
Métriques
HDR-PSNR
LPIPS
SSIM
Résultats
Résultats de performance de divers modèles sur ce benchmark
Nom du modèle | HDR-PSNR | LPIPS | SSIM | Paper Title | Repository |
|---|---|---|---|---|---|
| PhySG | 21.81 | 0.055 | 0.960 | PhySG: Inverse Rendering with Spherical Gaussians for Physics-based Material Editing and Relighting | - |
| Neural-PBIR | 26.01 | 0.023 | 0.979 | Neural-PBIR Reconstruction of Shape, Material, and Illumination | - |
| NeRFactor | 23.54 | 0.048 | 0.969 | NeRFactor: Neural Factorization of Shape and Reflectance Under an Unknown Illumination | |
| NeRD | 23.29 | 0.059 | 0.957 | NeRD: Neural Reflectance Decomposition from Image Collections | |
| InvRender | 23.76 | 0.046 | 0.970 | NeFII: Inverse Rendering for Reflectance Decomposition with Near-Field Indirect Illumination | |
| NVDiffRec | 22.91 | 0.039 | 0.963 | Extracting Triangular 3D Models, Materials, and Lighting From Images | |
| NVDiffRecMC | 24.43 | 0.036 | 0.972 | Shape, Light, and Material Decomposition from Images using Monte Carlo Rendering and Denoising |
0 of 7 row(s) selected.