Hybrid Training of Denoising Networks to Improve the Texture Acutance of Digital Cameras
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929319252590592 |
|---|---|
| author | Achddou, Raphaël Gousseau, Yann Ladjal, Saïd |
| author_facet | Achddou, Raphaël Gousseau, Yann Ladjal, Saïd |
| contents | In order to evaluate the capacity of a camera to render textures properly, the standard practice, used by classical scoring protocols, is to compute the frequential response to a dead leaves image target, from which is built a texture acutance metric. In this work, we propose a mixed training procedure for image restoration neural networks, relying on both natural and synthetic images, that yields a strong improvement of this acutance metric without impairing fidelity terms. The feasibility of the approach is demonstrated both on the denoising of RGB images and the full development of RAW images, opening the path to a systematic improvement of the texture acutance of real imaging devices. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_07212 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Hybrid Training of Denoising Networks to Improve the Texture Acutance of Digital Cameras Achddou, Raphaël Gousseau, Yann Ladjal, Saïd Image and Video Processing Artificial Intelligence Computer Vision and Pattern Recognition In order to evaluate the capacity of a camera to render textures properly, the standard practice, used by classical scoring protocols, is to compute the frequential response to a dead leaves image target, from which is built a texture acutance metric. In this work, we propose a mixed training procedure for image restoration neural networks, relying on both natural and synthetic images, that yields a strong improvement of this acutance metric without impairing fidelity terms. The feasibility of the approach is demonstrated both on the denoising of RGB images and the full development of RAW images, opening the path to a systematic improvement of the texture acutance of real imaging devices. |
| title | Hybrid Training of Denoising Networks to Improve the Texture Acutance of Digital Cameras |
| topic | Image and Video Processing Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2404.07212 |