Analog Video Restoration On Tape
Métriques
LPIPS
PSNR
SSIM
VMAF
Résultats
Résultats de performance de divers modèles sur ce benchmark
Nom du modèle | LPIPS | PSNR | SSIM | VMAF | Paper Title | Repository |
|---|---|---|---|---|---|---|
| BasicVSR++ | 0.098 | 31.66 | 0.916 | 78.91 | BasicVSR++: Improving Video Super-Resolution with Enhanced Propagation and Alignment | |
| RTN | 0.100 | 31.46 | 0.905 | 56.76 | Bringing Old Films Back to Life | |
| RVRT | 0.117 | 32.47 | 0.896 | 72.41 | Recurrent Video Restoration Transformer with Guided Deformable Attention | |
| TAPE | 0.052 | 35.53 | 0.946 | 83.61 | Reference-based Restoration of Digitized Analog Videotapes | |
| Agnolucci et al. | 0.060 | 34.96 | 0.940 | 77.83 | Restoration of Analog Videos Using Swin-UNet | |
| MemDeblur | 0.106 | 33.22 | 0.911 | 71.55 | Multi-Scale Memory-Based Video Deblurring | |
| MANA | 0.206 | 27.81 | 0.843 | 40.28 | Memory-Augmented Non-Local Attention for Video Super-Resolution |
0 of 7 row(s) selected.