AI Paper Atlas

Research map / Image generation / Image restoration

ShiftLUT: Spatial Shift Enhanced Look-Up Tables for Efficient Image Restoration

Xiaolong Zeng, Yitong Yu, Shiyao Xiong, Jinhua Hao, Ming Sun, Chao Zhou, Bin Wang

CVPR 2026 · Image generation › Image restoration · approach: Efficient restoration architectures · 0 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 Learnable Spatial Shift (LSS) module (ablation)ShiftLUT with varying Shift-Block counts and channel numbers · Set5 x4PSNR (dB): +0.37 to +0.46 across block counts, +0.30 to +0.43 across channelsImproved
add LSS to widen receptive field (ablation)ShiftLUT · LAM visualization of 16x16 patchdetail index (DI): 3.71 -> 6.13Improved
make the LSB branch a single 3x3 conv (asymmetric dual branch) instead of a symmetric deep LSB branch (ablation)ShiftLUT · average of five x4 SR benchmarksPSNR / latency: PSNR 28.19/0.8016 -> 28.19/0.8014; latency 164ms -> 84msImproved
use Error-bounded Adaptive Sampling (eps 0.4) to compress LUTs (ablation)ShiftLUT-L · Set5 x4LUT size / PSNR: 217KB -> 104KB at PSNR 31.33 unchangedImproved
increase EAS tolerance from 0.4 to 0.8 (ablation)ShiftLUT-L · Set5 x4LUT size / PSNR: 104KB -> 54KB; PSNR 31.33 -> 31.30Mixed
use uniform sampling (step 2) instead of EAS (ablation)ShiftLUT-L · Set5 x4runtime (ms): 84 -> 213Hurt
use DDM compression instead of EAS (ablation)ShiftLUT-L · Set5 x4PSNR / size / runtime: PSNR 31.33 -> 31.26; size 172KB vs 104KB; runtime 98ms vs 84msHurt
assign more bits to the MSB branch (MSB/LSB split) (ablation)ShiftLUT · x4 SRPSNR and LUT storageMixed
use a more complex offset prediction network in LSS (ablation)ShiftLUT · x4 SRPSNRNo clear effect
replace learned continuous offsets with fixed integer shifts at inference (two-stage training) (ablation)ShiftLUT LSS · inferenceinference costImproved
use ShiftLUT-L instead of TinyLUT-F on x4 SRLUT-based SISR · Manga109 x4PSNR (dB): 28.83 -> 29.16Improved
use ShiftLUT-L instead of TinyLUT-F for denoisingLUT-based denoising · Set12, BSD68 sigma 15PSNR (dB): 32.22 -> 32.43; 31.20 -> 31.32Improved

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