Self-supervised learning for phase retrieval
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916980072644608 |
|---|---|
| author | Sechaud, Victor Abry, Patrice Jacques, Laurent Tachella, Julián |
| author_facet | Sechaud, Victor Abry, Patrice Jacques, Laurent Tachella, Julián |
| contents | In recent years, deep neural networks have emerged as a solution for inverse imaging problems. These networks are generally trained using pairs of images: one degraded and the other of high quality, the latter being called 'ground truth'. However, in medical and scientific imaging, the lack of fully sampled data limits supervised learning. Recent advances have made it possible to reconstruct images from measurement data alone, eliminating the need for references. However, these methods remain limited to linear problems, excluding non-linear problems such as phase retrieval. We propose a self-supervised method that overcomes this limitation in the case of phase retrieval by using the natural invariance of images to translations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_26203 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Self-supervised learning for phase retrieval Sechaud, Victor Abry, Patrice Jacques, Laurent Tachella, Julián Information Retrieval Machine Learning In recent years, deep neural networks have emerged as a solution for inverse imaging problems. These networks are generally trained using pairs of images: one degraded and the other of high quality, the latter being called 'ground truth'. However, in medical and scientific imaging, the lack of fully sampled data limits supervised learning. Recent advances have made it possible to reconstruct images from measurement data alone, eliminating the need for references. However, these methods remain limited to linear problems, excluding non-linear problems such as phase retrieval. We propose a self-supervised method that overcomes this limitation in the case of phase retrieval by using the natural invariance of images to translations. |
| title | Self-supervised learning for phase retrieval |
| topic | Information Retrieval Machine Learning |
| url | https://arxiv.org/abs/2509.26203 |