Deep Equilibrium Models for Snapshot Compressive Imaging
DOI:
https://doi.org/10.1609/aaai.v37i3.25475Keywords:
CV: Computational Photography, Image & Video SynthesisAbstract
The ability of snapshot compressive imaging (SCI) systems to efficiently capture high-dimensional (HD) data has led to an inverse problem, which consists of recovering the HD signal from the compressed and noisy measurement. While reconstruction algorithms grow fast to solve it with the recent advances of deep learning, the fundamental issue of accurate and stable recovery remains. To this end, we propose deep equilibrium models (DEQ) for video SCI, fusing data-driven regularization and stable convergence in a theoretically sound manner. Each equilibrium model implicitly learns a nonexpansive operator and analytically computes the fixed point, thus enabling unlimited iterative steps and infinite network depth with only a constant memory requirement in training and testing. Specifically, we demonstrate how DEQ can be applied to two existing models for video SCI reconstruction: recurrent neural networks (RNN) and Plug-and-Play (PnP) algorithms. On a variety of datasets and real data, both quantitative and qualitative evaluations of our results demonstrate the effectiveness and stability of our proposed method. The code and models are available at: https://github.com/IndigoPurple/DEQSCI.Downloads
Published
2023-06-26
How to Cite
Zhao, Y., Zheng, S., & Yuan, X. (2023). Deep Equilibrium Models for Snapshot Compressive Imaging. Proceedings of the AAAI Conference on Artificial Intelligence, 37(3), 3642-3650. https://doi.org/10.1609/aaai.v37i3.25475
Issue
Section
AAAI Technical Track on Computer Vision III