AI Paper Atlas

Research map / Image generation / Image restoration

Multispectral Illumination Estimation Using Deep Unrolling Network

Yuqi Li, Qiang Fu, Wolfgang Heidrich

ICCV 2021 · Image generation › Image restoration · approach: Pansharpening and spectral fusion · 57 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
make unrolled-stage parameters trainable instead of fixed values (ablation)Unrolling illumination estimation network · Own dataset, multiple illuminationabsolute error dS: 4.23 -> 2.84Improved
add TV regularization block (ablation)Unrolling illumination estimation network · Own dataset, multiple illuminationabsolute error dS: 4.35 -> 2.84Improved
add low-rank regularization block (ablation)Unrolling illumination estimation network · Own dataset, multiple illuminationabsolute error dS: 4.21 -> 2.84Improved
add non-local self-attention transformer block (ablation)Unrolling illumination estimation network · Own dataset, multiple illuminationabsolute error dS: 3.63 -> 2.84Improved
add non-local self-attention transformer block (ablation)Unrolling illumination estimation network · Own dataset, multiple illuminationangular error dA: 0.25 -> 0.21Improved
use transformer instead of residual conv block for reflectance denoising (ablation)Unrolling illumination estimation network · Own dataset, multiple illuminationabsolute error dS: 3.50 -> 2.84Improved
use unrolling network instead of GrayEdge (best baseline)Unrolling network · Own dataset, single global illuminationabsolute error dS: 2.29 -> 1.80Improved
use unrolling network instead of PWIR (CNN) on unseen CAVE datasetUnrolling network · CAVE, single global illuminationabsolute error dS: 4.62 -> 1.87Improved
use unrolling network instead of AngularGANUnrolling network · Own dataset, multiple illuminationabsolute error dS: 3.77 -> 2.84Improved
use unrolling network instead of BUTDUnrolling network · CAVE, multiple illuminationangular error dA: 0.48 -> 0.31Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy