ADMM-DAD net: a deep unfolding network for analysis compressed sensing
This work addresses signal reconstruction for applications like image and speech processing, but it is incremental as it builds on existing deep unfolding methods with a specific algorithmic adaptation.
The paper tackled the problem of signal reconstruction in analysis compressed sensing by proposing a deep unfolding network based on the ADMM algorithm, which jointly learns a redundant analysis operator and reconstructs signals, outperforming a state-of-the-art unfolded ISTA decoder on both image and speech datasets.
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.