AI Paper Atlas

Research map / Image generation / Image restoration

Joint RGB-Spectral Decomposition Model Guided Image Enhancement in Mobile Photography

Kailai Zhou, Lijing Cai, Yibo Wang, Mengya Zhang, Bihan Wen, Qiu Shen, Xun Cao

ECCV 2024 · Image generation › Image restoration · approach: Pansharpening and spectral fusion · 1 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
add ideal shading prior S* to HDRNet (ablation)HDRNet · Mobile-Spec testPSNR: 27.75 -> 28.68Improved
add reflectance prior from low-res multispectral image (all 10 channels) (ablation)HDRNet + S* · Mobile-Spec testPSNR: 28.68 -> 29.68Improved
add material semantic prior with grid experts (ablation)HDRNet + S* + R* · Mobile-Spec testPSNR: 29.68 -> 30.14Improved
add Lr-MSI branch to joint decomposition model for material segmentation (ablation)joint decomposition model · Mobile-Spec material segmentationmIoU: 71.86 -> 78.93Improved
increase number of material categories from 1 to 6 (ablation)HDRNet with M* · Mobile-Spec testPSNR: 27.75 -> 29.11Improved
use two material categories instead of one (ablation)HDRNet with M* · Mobile-Spec testDelta E*: 5.12 -> 5.35Hurt
increase Lr-MSI spatial resolution from 16x16 to 256x256 (ablation)reflectance prior · Mobile-Spec testPSNR: 29.68 -> 29.78No clear effect
use visible-only vs all spectral channels for reflectance prior (ablation)reflectance prior · Mobile-Spec testPSNR: 29.24 -> 29.68Improved
use predicted priors from joint decomposition instead of ideal priors (ablation)JDM-HDRNet · Mobile-Spec testPSNR: 30.14 -> 29.83Mixed
add predicted material prior on top of predicted S and R (ablation)JDM-HDRNet · Mobile-Spec testPSNR: 29.57 -> 29.83Improved

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