白平衡优先，后调整：通过视觉语言模型评估的跨相机颜色恒常性
White-Balance First, Adjust Later: Cross-Camera Color Constancy via Vision-Language Evaluation
将颜色恒常性重新定义为迭代感知反馈问题，利用LoRA微调的VLM评估白平衡后的图像并给出残差色偏方向，替代直接RGB回归，为跨相机颜色恒常性提供新思路。
原文依据 We propose VLM-CC, a feedback-guided framework that formulates color constancy as an iterative refinement process.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论颜色恒常性旨在使物体颜色在不同光照下保持一致。跨相机泛化在颜色恒常性中仍具挑战性，因为基于学习的模型往往过拟合于训练相机的颜色响应特性，导致在其他相机拍摄的图像上性能下降。我们提出VLM-CC，一个反馈引导的框架，将颜色恒常性表述为迭代精炼过程。VLM-CC不直接从原始输入估计光源，而是执行由基于视觉语言模型（VLM）的评估驱动的迭代校正。在每次迭代中，使用当前估计对图像进行白平衡并转换为伪sRGB。一个轻量级LoRA微调的VLM随后评估校正后的图像，识别主要的残余色偏并提供定性反馈。该反馈被映射为残余光照方向（红、绿或蓝），并用于更新光源估计直至收敛。我们的核心思想是将颜色恒常性重新定义为迭代感知反馈问题，利用VLM评估而非直接RGB回归。通过用VLM引导的感知反馈替代直接RGB估计，VLM-CC在多个数据集上实现了跨相机颜色恒常性的最优鲁棒性。
英文原文摘要 1311 字符
Color constancy aims to keep object colors consistent under varying illumination. Cross-camera generalization in color constancy remains challenging because learning-based models often overfit to the color response characteristics of the training camera, resulting in degraded performance on images captured by other cameras. We propose VLM-CC, a feedback-guided framework that formulates color constancy as an iterative refinement process. Instead of directly estimating the illuminant from raw input, VLM-CC performs iterative correction driven by vision-language model (VLM)-based evaluation. At each iteration, the image is white-balanced using the current estimate and converted to pseudo-sRGB. A lightweight LoRA-tuned VLM then assesses the corrected image, identifying the dominant residual color cast and providing qualitative feedback. This feedback is mapped to a residual illumination direction (red, green, or blue) and used to update the illuminant estimate until convergence. Our key idea is to reframe color constancy as an iterative perceptual feedback problem, leveraging VLM evaluation instead of direct RGB regression. By replacing direct RGB estimation with VLM-guided perceptual feedback, VLM-CC achieves state-of-the-art robustness in cross-camera color constancy across multiple datasets.