Paint-Anything:图像生成与编辑中统一任意颜色控制
Paint-Anything: Unified Any-Color Control for Image Generation and Editing
朋友,DeepSeek 新出的这个 Paint-Anything 工具很厉害,能让你在生成和编辑图片时指定任意十六进制颜色,比之前的方法效果提升很大。
这篇论文提出了一种名为 Paint-Anything 的新方法,用于图像生成和编辑中的任意颜色控制。该方法通过对象级别的颜色监督学习一个共享的十六进制色码提示接口。
它构建了一个包含 50 万张图像的数据集 Paint-500K,并引入了 Any Color Benchmark (ACBench) 来衡量对象级十六进制颜色保真度。在 FLUX.
2-4B 模型上,该方法将 ACBench-T2I 和 ACBench-Edit 的分数分别提升了 85.3% 和 28.
3%。
Paint-Anything: Unified Any-Color Control for Image Generation and Editing
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.
本条情报归属 AITOP「论文」核心栏目,重点聚焦北美前沿实验室与硅谷科技巨头。在跨区域产业观察坐标中,该动态直接关联「AI 开发者工具链、Coding Agent 与工程化实践」研究主线,反映出当前技术演进与商业落地的最新风向。涉及核心实体与技术议题包括:Paint-Anything、图像生成、图像编辑、提示词工程,读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。
信源核验说明:本文原始线索采编自「arXiv cs.AI」(@Ji Xie, Dewei Zhou, Xinyu Huang, Zhennan Chen, Xun Wang),经 AITOP 多维评分管线结构化提炼。