无需训练即可精准保持图像局部色彩，适合设计类文本生成。
Inference-Time Loss-Guided Colour Preservation in Diffusion Sampling
- 用区域掩码和梯度引导，在推理阶段精确控制指定区域颜色。
- 通过分布感知损失函数，同时约束平均色与色差极端值，避免局部出错。
- 可直接接入Stable Diffusion绘图流程，适合需要精确配色的设计场景。
文本到图像扩散模型在色彩控制上仍存在明显缺陷，尤其在需严格匹配用户指定色彩的设计任务中。本文提出一种无需额外训练的推理阶段区域色彩保持方法。该方法结合三点：(i) 基于感兴趣区域（ROI）的图像修复实现空间选择性；(ii) 背景潜在表示重置，防止ROI外颜色漂移；(iii) 采用基于CIE Lab和线性RGB的复合损失函数，通过梯度引导进行潜在空间调整。损失函数引入CVaR风格与软最大值惩罚，不仅控制ROI平均颜色，还抑制像素级误差分布的尾部异常。通过延迟启动门控和时间依赖调度策略，确保各去噪步骤的引导稳定性。实验表明，仅控制均值的基线虽能满足平均色要求，但会产生显著的局部视觉错误，凸显分布感知目标的必要性。所提方法为精准色彩适配提供了一种实用、免训练的解决方案，可无缝集成至标准Stable Diffusion图像修复流程。
原文摘要 · Abstract (English)
Precise color control remains a persistent failure mode in text-to-image diffusion systems, particularly in design-oriented workflows where outputs must satisfy explicit, user-specified color targets. We present an inference-time, region-constrained color preservation method that steers a pretrained diffusion model without any additional training. Our approach combines (i) ROI-based inpainting for spatial selectivity, (ii) background-latent re-imposition to prevent color drift outside the ROI, and (iii) latent nudging via gradient guidance using a composite loss defined in CIE Lab and linear RGB. The loss is constructed to control not only the mean ROI color but also the tail of the pixelwise error distribution through CVaR-style and soft-maximum penalties, with a late-start gate and a time-dependent schedule to stabilize guidance across denoising steps. We show that mean-only baselines can satisfy average color constraints while producing perceptually salient local failures, motivating our distribution-aware objective. The resulting method provides a practical, training-free mechanism for targeted color adherence that can be integrated into standard Stable Diffusion inpainting pipelines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。