English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

This paper endeavors to advance the precision of snapshot compressive imaging(SCI) reconstruction for multispectral image (MSI). To achieve this, weintegrate the advantageous attributes of established SCI techniques and animage generative model, propose a novel structured zero-shot diffusion model,dubbed DiffSCI. DiffSCI leverages the structural insights from the deep priorand optimization-based methodologies, complemented by the generativecapabilities offered by the contemporary denoising diffusion model.Specifically, firstly, we employ a pre-trained diffusion model, which has beentrained on a substantial corpus of RGB images, as the generative denoiserwithin the Plug-and-Play framework for the first time. This integration allowsfor the successful completion of SCI reconstruction, especially in the casethat current methods struggle to address effectively. Secondly, wesystematically account for spectral band correlations and introduce a robustmethodology to mitigate wavelength mismatch, thus enabling seamless adaptationof the RGB diffusion model to MSIs. Thirdly, an accelerated algorithm isimplemented to expedite the resolution of the data subproblem. Thisaugmentation not only accelerates the convergence rate but also elevates thequality of the reconstruction process. We present extensive testing to showthat DiffSCI exhibits discernible performance enhancements over prevailingself-supervised and zero-shot approaches, surpassing even supervisedtransformer counterparts across both simulated and real datasets. Our code willbe available.

DiffSCI: Zero-Shot Snapshot Compressive Imaging via Iterative Spectral Diffusion Model