DiffFPR: Diffusion Prior for Oversampled Fourier Phase Retrieval

Ji Li, Chao Wang
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:28673-28687, 2024.

Abstract

This paper tackled the challenging Fourier phase retrieval problem, the absolute uniqueness of which does not hold. The existence of equivalent solution (a.k.a. trivial solution ambiguity) hinders the successful recovery, especially for multi-channel color image. The traditional iterative engine, such as the Relaxed Averaged Alternating Reflections (RAAR), can be applied to reconstruct the image channel-wisely. However, due to the relative uniqueness of the solution, the restoration is not automatically aligned with the accurate orientation for each channel, resulting in a reconstructed image that deviates significantly from the true solution manifold. To address this issue, by penalizing the mismatch of the image channels, a diffusion model as the strong prior of the color image is integrated into the iterative engine. The combination of the traditional iterative engine and the diffusion model provides an effective solution to the oversampled Fourier phase retrieval. The formed algorithm, DiffFPR, is validated by experiments. The code is available at https://github.com/Chilie/DiffFPR.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-li24bj, title = {{D}iff{FPR}: Diffusion Prior for Oversampled {F}ourier Phase Retrieval}, author = {Li, Ji and Wang, Chao}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {28673--28687}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/li24bj/li24bj.pdf}, url = {https://proceedings.mlr.press/v235/li24bj.html}, abstract = {This paper tackled the challenging Fourier phase retrieval problem, the absolute uniqueness of which does not hold. The existence of equivalent solution (a.k.a. trivial solution ambiguity) hinders the successful recovery, especially for multi-channel color image. The traditional iterative engine, such as the Relaxed Averaged Alternating Reflections (RAAR), can be applied to reconstruct the image channel-wisely. However, due to the relative uniqueness of the solution, the restoration is not automatically aligned with the accurate orientation for each channel, resulting in a reconstructed image that deviates significantly from the true solution manifold. To address this issue, by penalizing the mismatch of the image channels, a diffusion model as the strong prior of the color image is integrated into the iterative engine. The combination of the traditional iterative engine and the diffusion model provides an effective solution to the oversampled Fourier phase retrieval. The formed algorithm, DiffFPR, is validated by experiments. The code is available at https://github.com/Chilie/DiffFPR.} }
Endnote
%0 Conference Paper %T DiffFPR: Diffusion Prior for Oversampled Fourier Phase Retrieval %A Ji Li %A Chao Wang %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-li24bj %I PMLR %P 28673--28687 %U https://proceedings.mlr.press/v235/li24bj.html %V 235 %X This paper tackled the challenging Fourier phase retrieval problem, the absolute uniqueness of which does not hold. The existence of equivalent solution (a.k.a. trivial solution ambiguity) hinders the successful recovery, especially for multi-channel color image. The traditional iterative engine, such as the Relaxed Averaged Alternating Reflections (RAAR), can be applied to reconstruct the image channel-wisely. However, due to the relative uniqueness of the solution, the restoration is not automatically aligned with the accurate orientation for each channel, resulting in a reconstructed image that deviates significantly from the true solution manifold. To address this issue, by penalizing the mismatch of the image channels, a diffusion model as the strong prior of the color image is integrated into the iterative engine. The combination of the traditional iterative engine and the diffusion model provides an effective solution to the oversampled Fourier phase retrieval. The formed algorithm, DiffFPR, is validated by experiments. The code is available at https://github.com/Chilie/DiffFPR.
APA
Li, J. & Wang, C.. (2024). DiffFPR: Diffusion Prior for Oversampled Fourier Phase Retrieval. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:28673-28687 Available from https://proceedings.mlr.press/v235/li24bj.html.

Related Material