JR2net: A Joint Non-Linear Representation and Recovery Network for Compressive Spectral Imaging

TL;DR

JR2net introduces a unified deep learning framework that jointly learns non-linear spectral image representations and performs recovery, significantly improving accuracy and speed over existing methods in compressive spectral imaging.

Contribution

It proposes a novel joint optimization network that integrates representation learning and image recovery for CSI, unlike previous detached approaches.

Findings

Achieves up to 2.57 dB PSNR improvement.

Runs approximately 2000 times faster.

Outperforms state-of-the-art methods in spectral image recovery.

Abstract

Deep learning models are state-of-the-art in compressive spectral imaging (CSI) recovery. These methods use a deep neural network (DNN) as an image generator to learn non-linear mapping from compressed measurements to the spectral image. For instance, the deep spectral prior approach uses a convolutional autoencoder network (CAE) in the optimization algorithm to recover the spectral image by using a non-linear representation. However, the CAE training is detached from the recovery problem, which does not guarantee optimal representation of the spectral images for the CSI problem. This work proposes a joint non-linear representation and recovery network (JR2net), linking the representation and recovery task into a single optimization problem. JR2net consists of an optimization-inspired network following an ADMM formulation that learns a non-linear low-dimensional representation and simultaneously performs the spectral image recovery, trained via the end-to-end approach. Experimental results show the superiority of the proposed method with improvements up to 2.57 dB in PSNR and performance around 2000 times faster than state-of-the-art methods.