| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Image Quality Assessment Correlation | RealEstate10K | PLCC1 | 52 | |
| Novel View Synthesis | RealEstate-10K 2-view | PSNR30.3 | 32 | |
| Novel View Synthesis | RealEstate 10k (RE10k) (test) | PSNR27.949 | 24 | |
| Novel View Synthesis | RealEstate-10K 2 views (test) | LPIPS0.0635 | 15 | |
| Camera Controllability | RealEstate10K (test) | mRotErr1.097 | 10 | |
| Novel View Synthesis | RealEstate10K 58 (test) | PSNR16.362 | 8 | |
| Multi-view generation | RealEstate10K | MEt3R0.12 | 7 | |
| Novel View Synthesis and Depth Estimation | RealEstate-10K | LPIPS0.0706 | 6 | |
| Image-to-Video Generation | RealEstate 122 (test) | PSNR18.58 | 6 | |
| Camera-controlled Video Generation | RealEstate dataset | FVD54.83 | 6 | |
| Novel View Synthesis | RealEstate10K Long-range source-target pairs | FID3 | 6 | |
| Novel View Synthesis | RealEstate10K Mid-range source-target pairs | FID2.58 | 6 | |
| Translation estimation | RealEstate-10K (Avg) | Avg Translation Error (m)0.332 | 6 | |
| Translation estimation | RealEstate-10K (Medium overlap) | Avg Translation Error (m)0.203 | 6 | |
| Translation estimation | RealEstate-10K (Small overlap) | Avg Translation Error (m)0.532 | 6 | |
| Perpetual view generation | RealEstate-10K | PSNR23.52 | 5 | |
| Novel View Synthesis | RealEstate10K Mid-range (test) | FID2.58 | 5 | |
| Novel view rendering | RealEstate-10K (large overlap) | PSNR26.199 | 5 | |
| Depth Estimation | RealEstate-10K 2 input views | MAE0.3269 | 4 | |
| Novel View Synthesis | RealEstate10k | User Preference Score0.6391 | 2 |