An interpretable and flexible fusion prior to boost hyperspectral imaging reconstruction

Wei He, Zongliang Wu, Naoto Yokoya, Xin Yuan

Abstract: Highlights•Propose a hyperspectral fusion theory for HSI imaging reconstruction.•HSI imaging reconstruction is distilled to that of subspace reconstruction.•The subspace distillation prior (SP) can improve any deep learning models.•Experimental results show advantage of hyperspectral fusion theory.