• Corpus ID: 211572964

Class-Specific Blind Deconvolutional Phase Retrieval Under a Generative Prior

@article{Shamshad2020ClassSpecificBD,
  title={Class-Specific Blind Deconvolutional Phase Retrieval Under a Generative Prior},
  author={Fahad Shamshad and Ali Ahmed},
  journal={ArXiv},
  year={2020},
  volume={abs/2002.12578}
}
In this paper, we consider the highly ill-posed problem of jointly recovering two real-valued signals from the phaseless measurements of their circular convolution. The problem arises in various imaging modalities such as Fourier ptychography, X-ray crystallography, and in visible light communication. We propose to solve this inverse problem using alternating gradient descent algorithm under two pretrained deep generative networks as priors; one is trained on sharp images and the other on blur… 

Figures and Tables from this paper

References

SHOWING 1-10 OF 29 REFERENCES

prDeep: Robust Phase Retrieval with a Flexible Deep Network

This work uses the regularization-by-denoising framework and a convolutional neural network denoiser to create prDeep, a new phase retrieval algorithm that is both robust and broadly applicable and test and validate in simulation to demonstrate that it is robust to noise and can handle a variety of system models.

Robust Compressive Phase Retrieval via Deep Generative Priors

This paper proposes a new framework to regularize the highly ill-posed and non-linear phase retrieval problem through deep generative priors using simple gradient descent algorithm and demonstrates effectiveness of proposed algorithm for random Gaussian measurements and Fourier friendly measurements.

Simultaneous Blind Deconvolution and Phase Retrieval with Tensor Iterative Hard Thresholding

  • S. LiGongguo TangM. Wakin
  • Computer Science
    ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • 2019
This work shows that this non-linear problem can be reformulated as a low-rank tensor recovery problem and proposes an algorithm named TIHT-BDPR to recover the unknown parameters.

Image Deblurring with a Class-Specific Prior

It is shown that the subspace of band-pass filtered images and their intensity distributions serve as useful priors for recovering image frequencies that are difficult to recover by generic image priors.

Deep Image Prior

It is shown that a randomly-initialized neural network can be used as a handcrafted prior with excellent results in standard inverse problems such as denoising, super-resolution, and inpainting.

Adaptive Ptych: Leveraging Image Adaptive Generative Priors for Subsampled Fourier Ptychography

This work proposes to make pretrained generator image adaptive by modifying it to better represent a single image (at test time) that is consistent with the subsampled FP measurements, demonstrating the superiority of the proposed approach over recent subsampling FP methods in terms of both quantitative metrics and visual quality.

Blind Deconvolutional Phase Retrieval via Convex Programming

This work proves that if the two signals belong to known random subspaces of dimensions $k$ and $n$, then they can be recovered up to the inherent scaling ambiguity with $m >> (k+n) \log^2 m$ phaseless measurements.

Simultaneous Phase Retrieval and Blind Deconvolution via Convex Programming

It is proved that if the two signals belong to known random subspaces of dimensions $k$ and $n$, then they can be recovered up to the inherent scaling ambiguity with $m \gg (k+n) \log^2 m$ phaseless measurements.

Phase Retrieval Under a Generative Prior

This paper proposes a novel framework for phase retrieval by modeling natural signals as being in the range of a deep generative neural network and enforcing this prior directly by optimizing an empirical risk objective over the domain of the generator, and confirms that exploiting generative models in phase retrieval tasks outperforms sparse phase retrieval methods.

Invertible generative models for inverse problems: mitigating representation error and dataset bias

It is demonstrated that invertible neural networks, which have zero representation error by design, can be effective natural signal priors at inverse problems such as denoising, compressive sensing, and inpainting.