ADMM-DAD net: a deep unfolding network for analysis compressed sensing

TL;DR

This paper introduces ADMM-DAD net, a deep unfolding neural network based on ADMM for analysis compressed sensing, which jointly learns sparsification and signal reconstruction, outperforming existing methods on image and speech data.

Contribution

It presents a novel deep unfolding network that combines ADMM with learned analysis operators, improving reconstruction performance over prior state-of-the-art methods.

Findings

Outperforms existing deep unfolding networks on image datasets

Achieves superior results on speech datasets

Demonstrates robustness across different data types

Abstract

In this paper, we propose a new deep unfolding neural network based on the ADMM algorithm for analysis Compressed Sensing. The proposed network jointly learns a redundant analysis operator for sparsification and reconstructs the signal of interest. We compare our proposed network with a state-of-the-art unfolded ISTA decoder, that also learns an orthogonal sparsifier. Moreover, we consider not only image, but also speech datasets as test examples. Computational experiments demonstrate that our proposed network outperforms the state-of-the-art deep unfolding network, consistently for both real-world image and speech datasets.