AI Paper Atlas

Research map / Image generation / Editing and style transfer

StatLUT: Statistical Feature-Driven Multimodal 3D LUT Generation for Photorealistic Style Transfer

Yifan Wang, Zhixiang Hao, Yu Wang, Congchao Zhu

NeurIPS 2026 · Image generation › Editing and style transfer · approach: Arbitrary neural style transfer

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes StatLUT, a method that generates 3D lookup tables for photorealistic style transfer from statistical features of multimodal inputs.

Problem

Photorealistic style transfer must match a reference's color and tone while preserving content structure; existing deep methods suffer from semantic entanglement from pre-trained encoders causing spatial distortions, and pixel-level mappings ignore color gamut topology.

Approach

Generates 3D LUTs driven by statistical features rather than semantic encoder features, avoiding semantic entanglement and enforcing color gamut structure; the abstract is truncated so further details are not stated.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Replace RGB 1D histograms with Lab L, ab and ab-conditioned mean luminance features (ablation)StatLUT · Feature ablationDist. to Ideal: 0.286 -> 0.281Improved
Add ab-conditioned mean luminance to Lab L and ab histograms (ablation)StatLUT · Feature ablationDist. to Ideal: 0.304 -> 0.281Improved
Use Lab Lab histograms without abL vs RGB histograms (ablation)StatLUT · Feature ablationStyle Sim.: 0.874 -> 0.848Hurt
Use MR-Mapper instead of simple MLP (ablation)StatLUT · Mapper ablationDist. to Ideal: 0.286 -> 0.259Improved
Use MR-Mapper instead of simple MLP (ablation)StatLUT · Mapper ablationparameters (M): 428.97 -> 0.38Improved
Use MR-Mapper instead of bottleneck MLP (ablation)StatLUT · Mapper ablationContent Sim.: 0.726 -> 0.760Improved
Use MR-Mapper instead of bottleneck MLP (ablation)StatLUT · Mapper ablationMACs (M): 5.24 -> 13.82Hurt
Increase LUT size from 16 to 20 or 24 (ablation)StatLUT · LUT size ablationDist. to Ideal: 0.259 -> 0.268 / 0.260No clear effect
Increase LUT size from 8 to 16 (ablation)StatLUT · LUT size ablationDist. to Ideal: 0.322 -> 0.259Improved
Use text-driven H-Diffuser features instead of reference imageStatLUT (Text) · NASStyle Sim.: 0.746 -> 0.498Hurt
Use StatLUT image-driven vs NLUTStatLUT · NASContent Sim.: 0.640 -> 0.707Improved
Use StatLUT vs Neural Preset on content similarityStatLUT · NASContent Sim.: 0.709 -> 0.707Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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