Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2501.18545 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913973850341376 |
|---|---|
| author | Ahn, Kyusu Kim, JiSoo Lee, Sangik Lee, HyunGyu Ko, Byeonghyun Park, Chanwoo Lee, Jaejin |
| author_facet | Ahn, Kyusu Kim, JiSoo Lee, Sangik Lee, HyunGyu Ko, Byeonghyun Park, Chanwoo Lee, Jaejin |
| contents | Even though an Under-Display Camera (UDC) is an advanced imaging system, the display panel significantly degrades captured images or videos, introducing low transmittance, blur, noise, and flare issues. Tackling such issues is challenging because of the complex degradation of UDCs, including diverse flare patterns. However, no dataset contains videos of real-world UDC degradation. In this paper, we propose a real-world UDC video dataset called UDC-VIT. Unlike existing datasets, UDC-VIT exclusively includes human motions for facial recognition. We propose a video-capturing system to acquire clean and UDC-degraded videos of the same scene simultaneously. Then, we align a pair of captured videos frame by frame, using discrete Fourier transform (DFT). We compare UDC-VIT with six representative UDC still image datasets and two existing UDC video datasets. Using six deep-learning models, we compare UDC-VIT and an existing synthetic UDC video dataset. The results indicate the ineffectiveness of models trained on earlier synthetic UDC video datasets, as they do not reflect the actual characteristics of UDC-degraded videos. We also demonstrate the importance of effective UDC restoration by evaluating face recognition accuracy concerning PSNR, SSIM, and LPIPS scores. UDC-VIT is available at our official GitHub repository. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_18545 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | UDC-VIT: A Real-World Video Dataset for Under-Display Cameras Ahn, Kyusu Kim, JiSoo Lee, Sangik Lee, HyunGyu Ko, Byeonghyun Park, Chanwoo Lee, Jaejin Computer Vision and Pattern Recognition Even though an Under-Display Camera (UDC) is an advanced imaging system, the display panel significantly degrades captured images or videos, introducing low transmittance, blur, noise, and flare issues. Tackling such issues is challenging because of the complex degradation of UDCs, including diverse flare patterns. However, no dataset contains videos of real-world UDC degradation. In this paper, we propose a real-world UDC video dataset called UDC-VIT. Unlike existing datasets, UDC-VIT exclusively includes human motions for facial recognition. We propose a video-capturing system to acquire clean and UDC-degraded videos of the same scene simultaneously. Then, we align a pair of captured videos frame by frame, using discrete Fourier transform (DFT). We compare UDC-VIT with six representative UDC still image datasets and two existing UDC video datasets. Using six deep-learning models, we compare UDC-VIT and an existing synthetic UDC video dataset. The results indicate the ineffectiveness of models trained on earlier synthetic UDC video datasets, as they do not reflect the actual characteristics of UDC-degraded videos. We also demonstrate the importance of effective UDC restoration by evaluating face recognition accuracy concerning PSNR, SSIM, and LPIPS scores. UDC-VIT is available at our official GitHub repository. |
| title | UDC-VIT: A Real-World Video Dataset for Under-Display Cameras |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2501.18545 |