Our unfolding method comprises two cru- cial components: the Dual Prior Framework (DPF) and Fo- cused Attention (FA). DPF, beyond the typical image prior,.
scholar.google.com › citations
Our work has been accepted by CVPR, codes and results are coming soon (July or August). The codes and pre-trained weights have been released.
Abstract: Recently, deep unfolding methods have achieved remarkable success in the realm of Snapshot Compressive Imaging (SCI) reconstruction.
We derive an effective Dual Prior Unfolding (DPU), which achieves the joint utilization of multiple deep priors and greatly improves iteration efficiency.
We introduce a Dual Prior Unfolding SCI reconstruction model, which achieves the joint utilization of multiple deep priors and greatly improves iteration ...
It can be intuitively observed that our DPU yields more detailed content, cleaner textures, and fewer artifacts than the other competing methods. Mean- while, ...
Sep 24, 2024 · DPU [50] implements an HSI reconstruction model based on dual prior unfolding, which improves iteration efficiency by jointly utilizing multiple ...
Publication. MambaSCI: Efficient Mamba-UNet for Quad-Bayer Patterned Video Snapshot Compressive Imaging ... Dual Prior Unfolding for Snapshot Compressive Imaging.
Improving Spectral Snapshot Reconstruction with Spectral-Spatial Rectification · Dual Prior Unfolding for Snapshot Compressive Imaging.
Aug 30, 2024 · This paper introduces the Degradation-Aware Deep Unfolding Network (DADUN). DADUN leverages estimated priors from compressed frames and the physical mask to ...