Enhanced deep unrolling networks for snapshot compressive hyperspectral imaging

Xinran Qin, Yuhui Quan, Hui Ji

Abstract: Highlights•An enhanced unrolling network for hyperspectral image (HSI) reconstruction.•A momentum-inspired memory-assistant module, modeling gradient descent steps.•Cross-stage self-attention for unrolling networks to model the proximal mapping.•A spectral geometry consistency loss to leverage specific characteristics of HSIs.