AceTone：连接文字与色彩的条件图像分级
AceTone: Bridging Words and Colors for Conditional Image Grading
将颜色分级建模为生成3D-LUT token序列的任务，利用VQ-VAE压缩LUT并用视觉语言模型预测，为图像编辑提供了一种可复用的、语言驱动的全局颜色变换范式。
原文依据 AceTone formulates grading as a generative color transformation task, where a model directly produces 3D-LUTs conditioned on text prompts or reference images.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论色彩影响我们对图像风格和情感的解读。以往的颜色分级方法依赖逐块重着色或固定滤波器组，难以泛化到不同的创作意图或与人类审美偏好对齐。本研究提出AceTone，这是首个在统一框架内支持多模态条件颜色分级的方法。AceTone将分级表述为生成式颜色变换任务，模型直接生成以文本提示或参考图像为条件的3D-LUT。我们开发了基于VQ-VAE的分词器，将3x32^3的LUT向量压缩为64个离散标记，保真度ΔE<2。我们进一步构建了大规模数据集AceTone-800K，并训练视觉语言模型预测LUT标记，随后通过强化学习使输出与感知保真度和美学对齐。实验表明，AceTone在文本引导和参考引导的分级任务上均达到最优性能，LPIPS相比现有方法提升高达50%。人类评估确认AceTone的结果视觉愉悦且风格连贯，展示了通向语言驱动、美学对齐颜色分级的新途径。项目页面：github.com/martian422/AceTone。
英文原文摘要 1268 字符
Color affects how we interpret image style and emotion. Previous color grading methods rely on patch-wise recoloring or fixed filter banks, struggling to generalize across creative intents or align with human aesthetic preferences. In this study, we propose **AceTone**, the first approach that supports multimodal conditioned color grading within a unified framework. AceTone formulates grading as a generative color transformation task, where a model directly produces 3D-LUTs conditioned on text prompts or reference images. We develop a VQ-VAE-based tokenizer which compresses a 3x32^3 LUT vector to 64 discrete tokens with \Delta \text E <2 fidelity. We further build a large-scale dataset, AceTone-800K, and train a vision-language model to predict LUT tokens, followed by reinforcement learning to align outputs with perceptual fidelity and aesthetics. Experiments show that AceTone achieves state-of-the-art performance on both text-guided and reference-guided grading tasks, improving LPIPS by up to **50%** over existing methods. Human evaluations confirm that AceTone's results are visually pleasing and stylistically coherent, demonstrating a new pathway toward language-driven, aesthetic-aligned color grading. Project Page: github.com/martian422/AceTone.