Momentum accelerated unfolding network with spectral-spatial prior for computational spectral imaging

Open Webpage

Zeyu Cai, Chunlu Li, Yi Yu, Chengqian Jin, Feipeng Da

Published: 01 Jan 2024, Last Modified: 06 Nov 2024Appl. Soft Comput. 2024EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•MAUN with learnable hyper-parameters and Momentum methods ensures better accuracy.•A reprojection loss is proposed to improve the reconstruction accuracy of HIS.•MAUNSS exceeds the previous SOTA methods while requiring less memory costs.