Name
Title | ||
|---|---|---|
GAN-based Projector for Faster Recovery in Compressed Sensing with Convergence Guarantees. |
Abstract | ||
|---|---|---|
A Generative Adversarial Network (GAN) with generator $G$ trained to model the prior of images has been shown to perform better than sparsity-based regularizers in ill-posed inverse problems. In this work, we propose a new method of deploying a GAN-based prior to solve linear inverse problems using projected gradient descent (PGD). Our method learns a network-based projector for use in the PGD algorithm, eliminating the need for expensive computation of the Jacobian of $G$. Experiments show that our approach provides a speed-up of $30text{-}40times$ over earlier GAN-based recovery methods for similar accuracy in compressed sensing. Our main theoretical result is that if the measurement matrix is moderately conditioned for range($G$) and the projector is $delta$-approximate, then the algorithm is guaranteed to reach $O(delta)$ reconstruction error in $O(log(1/delta))$ steps in the low noise regime. Additionally, we propose a fast method to design such measurement matrices for a given $G$. Extensive experiments demonstrate the efficacy of this method by requiring $5text{-}10times$ fewer measurements than random Gaussian measurement matrices for comparable recovery performance. |
Year | Venue | DocType |
|---|---|---|
2019 | arXiv: Learning | Journal |
Volume | Citations | PageRank |
abs/1902.09698 | 0 | 0.34 |
References | Authors | |
6 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Ankit Raj | 1 | 0 | 1.35 |
| Yuqi Li | 2 | 13 | 8.05 |
| Yoram Bresler | 3 | 1104 | 119.17 |