Despite extensive research on UDC images and their restoration models, studies on videos have yet to be significantly explored. While two UDC video datasets exist, they primarily focus on unrealistic or synthetic UDC degradation rather than real-world UDC degradation. In this paper, we propose a real-world UDC video dataset called . Unlike existing datasets, only exclusively includes human motions that target facial recognition.
Ideally, we would like to compare with two existing UDC video datasets, PexelsUDC and VidUDC33K. However, since PexelsUDC is not publicly available, we use the P-OLED dataset used to create it. Table below gives a summary of the eight previous UDC datasets.
| Dataset |
Type |
Scene |
Dataset size |
Resolution |
fps |
Flare |
Face |
Publicly available |
Publication |
| T/P-OLED |
Image |
Synthetic |
300 |
1024x2048x3 |
- |
|
|
✓ |
CVPR ‘21 |
| SYNTH |
Image |
Synthetic |
2,376 |
800x800x3 |
- |
✓ |
|
✓ |
CVPR ‘21 |
| Pseudo-real |
Image |
Real |
6,747 |
512x512x3 |
- |
✓ |
|
✓ |
CVPR ‘23 |
| UDC-SIT |
Image |
Real |
2,340 |
1792x1280x4 |
- |
✓ |
|
✓ |
NeurIPS ‘23 |
| Tan et al. |
Image |
Synthetic |
73,000 |
- |
- |
|
✓ |
|
TCSVT ‘23 |
| Wang et al. |
Image |
Synthetic |
56,126 |
- |
- |
|
✓ |
|
arXiv ‘24 |
| PexelsUDC-T/P |
Video |
Synthetic |
160x100 (16,000) |
1280x720x3 |
25-50 |
|
|
|
arXiv ‘23 |
| VidUDC33K |
Video |
Synthetic |
677x50 (33,850) |
1920x1080x3 |
- |
✓ |
|
✓ |
AAAI ‘24 |
| UDC-VIT |
Video |
Real |
647x180 (116,460) |
1900x1060x3 |
60 |
✓ |
✓ |
✓ |
|