Fast Hierarchical Deep Unfolding Network for Image Compressed Sensing

TL;DR

The paper introduces FHDUN, a fast hierarchical deep unfolding network for image compressed sensing that enhances expressiveness and adaptability, reduces iterations, and outperforms current methods.

Contribution

It proposes a novel hierarchical unfolding architecture with dynamic hyperparameter generation and acceleration, improving performance and efficiency in image compressed sensing.

Findings

FHDUN saves over 50% of iterative loops compared to recent DUNs.

FHDUN outperforms state-of-the-art CS methods in experiments.

The model adapts hyperparameters dynamically based on input content.

Abstract

By integrating certain optimization solvers with deep neural network, deep unfolding network (DUN) has attracted much attention in recent years for image compressed sensing (CS). However, there still exist several issues in existing DUNs: 1) For each iteration, a simple stacked convolutional network is usually adopted, which apparently limits the expressiveness of these models. 2) Once the training is completed, most hyperparameters of existing DUNs are fixed for any input content, which significantly weakens their adaptability. In this paper, by unfolding the Fast Iterative Shrinkage-Thresholding Algorithm (FISTA), a novel fast hierarchical DUN, dubbed FHDUN, is proposed for image compressed sensing, in which a well-designed hierarchical unfolding architecture is developed to cooperatively explore richer contextual prior information in multi-scale spaces. To further enhance the adaptability, series of hyperparametric generation networks are developed in our framework to dynamically produce the corresponding optimal hyperparameters according to the input content. Furthermore, due to the accelerated policy in FISTA, the newly embedded acceleration module makes the proposed FHDUN save more than 50% of the iterative loops against recent DUNs. Extensive CS experiments manifest that the proposed FHDUN outperforms existing state-of-the-art CS methods, while maintaining fewer iterations.