Paint-Anything: Unified Any-Color Control for Image Generation and Editing

Hugging Face Daily Papers Papers
image-generation color-control hex-prompt dataset benchmark research ai

Summary

Paint-Anything introduces a unified hex-prompt interface for color control in image generation and editing, trained on a custom dataset and evaluated on a new benchmark showing significant improvements in color fidelity.

Professional design requires any-color control: the ability to specify an object's target color with any 24-bit hex value for image generation and editing. Prior work has explored color generation, editing, and colorization, but often relies on dedicated color representations or specialized inference procedures. Advances in large language models offer a simpler starting point: even compact models can associate hex values with color semantics. We present Paint-Anything, which learns a shared hex-prompt interface for generation and editing through object-level color supervision. We develop a data pipeline that constructs Paint-500K from real images through object grounding, perceptual color labeling, and editing-pair synthesis. Since shadows make real-image labels only approximate colors, we complement this supervision with pure-color anchors whose pixels exactly match their paired hex values. These anchors are used only at high-noise timesteps, leaving low-noise training to natural images. We further introduce Any Color Benchmark (ACBench), comprising ACBench-T2I and ACBench-Edit, to measure object-level hex color fidelity across both tasks. On FLUX.2-4B, Paint-Anything improves ACBench-T2I and ACBench-Edit scores by 85.3% and 28.3%, respectively, relative to the base model, with ablations supporting the training recipe. It also achieves the highest average CompColor score among the compared methods.
Original Article
View Cached Full Text

Cached at: 09/21/26, 07:21 AM

Paper page - Paint-Anything: Unified Any-Color Control for Image Generation and Editing

Source: https://huggingface.co/papers/2609.20816 Published on Sep 17

·

Submitted byhttps://huggingface.co/sanaka87

Ji Xieon Sep 21

Abstract

Professionaldesignrequiresany-colorcontrol:theabilitytospecifyanobject’stargetcolorwithany24-bithexvalueforimagegenerationandediting.Priorworkhasexploredcolorgeneration,editing,andcolorization,butoftenreliesondedicatedcolorrepresentationsorspecializedinferenceprocedures.Advancesinlargelanguagemodelsofferasimplerstartingpoint:evencompactmodelscanassociatehexvalueswithcolorsemantics.WepresentPaint-Anything,whichlearnsasharedhex-promptinterfaceforgenerationandeditingthroughobject-levelcolorsupervision.WedevelopadatapipelinethatconstructsPaint-500Kfromrealimagesthroughobjectgrounding,perceptualcolorlabeling,andediting-pairsynthesis.Sinceshadowsmakereal-imagelabelsonlyapproximatecolors,wecomplementthissupervisionwithpure-coloranchorswhosepixelsexactlymatchtheirpairedhexvalues.Theseanchorsareusedonlyathigh-noisetimesteps,leavinglow-noisetrainingtonaturalimages.WefurtherintroduceAnyColorBenchmark(ACBench),comprisingACBench-T2IandACBench-Edit,tomeasureobject-levelhexcolorfidelityacrossbothtasks.OnFLUX.2-4B,Paint-AnythingimprovesACBench-T2IandACBench-Editscoresby85.3%and28.3%,respectively,relativetothebasemodel,withablationssupportingthetrainingrecipe.ItalsoachievesthehighestaverageCompColorscoreamongthecomparedmethods.

View arXiv pageView PDFAdd to collection

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2609.20816 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2609.20816 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2609.20816 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

PaintBench: Deterministic Evaluation of Precise Visual Editing

Hugging Face Daily Papers

PaintBench is a new benchmark for evaluating precise visual editing in multimodal models, covering 20 operations across 4 categories with deterministic pixel-level evaluation. Testing 11 models reveals overall low performance, with the best model scoring only 17.1% mIoU.

Mojave Paint

Product Hunt

Mojave Paint is a tool for direct manipulation of static images on the Mac platform.

prunaai/p-image-edit

Replicate Explore

Pruna's p-image-edit is a premium AI model on Replicate offering fast state-of-the-art image editing under one second, combining speed, affordability, and high visual quality with precise prompt adherence and text rendering capabilities.