VFHQ
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Video Face Restoration | VFHQ (test) | PSNR30.7289 | 25 | |
| Portrait Animation (Self-reenactment) | VFHQ (test) | FVD164.1895 | 23 | |
| Cross-identity Reenactment | VFHQ (test) | CSIM0.6757 | 23 | |
| Self-reenactment | VFHQ (test) | PSNR23.24 | 23 | |
| Talking Head Generation | VFHQ (test) | FID13.67 | 16 | |
| Blind Face Video Restoration | VFHQ (test) | PSNR29.563 | 14 | |
| Face Video Restoration | VFHQ (test) | PSNR26.231 | 11 | |
| Video Face Restoration | VFHQ milder degradation settings (test) | PSNR28.945 | 11 | |
| Face Video Restoration | VFHQ heavy degradation (test) | PSNR24.156 | 11 | |
| Face Reenactment | VFHQ Cross-reenactment zero-shot | CSIM0.675 | 11 | |
| Face Reenactment | VFHQ Self-reenactment one-shot | PSNR23.34 | 11 | |
| Neural Rendering Reenactment | VFHQ | FPS280 | 11 | |
| 3D Portrait Animation (Cross Reenactment) | VFHQ 1.0 (test) | CSIM69.4 | 11 | |
| Novel expression synthesis | VFHQ monocular (test) | PSNR23.35 | 10 | |
| Facial Reenactment | VFHQ Cross-Reenactment | ID-SIM0.892 | 8 | |
| Facial Reenactment | VFHQ Self-Reenactment | MSE0.018 | 8 | |
| Talking Head Generation | VFHQ | FID0 | 7 | |
| 3D Portrait Animation (Self Reenactment) | VFHQ 1.0 (test) | PSNR23.47 | 7 | |
| Lip-Syncing | VFHQ | FID9.12 | 7 | |
| Face Video Super-Resolution | VFHQ | PSNR26.35 | 6 | |
| Cross Reenactment | VFHQ | CSIM0.621 | 6 | |
| self-reenactment | VFHQ | LPIPS0.158 | 6 | |
| Cross-reenactment | VFHQ 512x512 | FID87 | 6 | |
| Portrait Animation | VFHQ (Cross-identity) | CSIM70.7 | 6 | |
| Portrait Animation | VFHQ Same-identity | PSNR23.555 | 6 |
Showing 25 of 35 rows