SAFE：纯彩色场景中基于学习色彩空间的色彩恒常性的场景感知特征调制
SAFE: Scene-Aware Feature Modulation for Color Constancy with Learned Color Space in Pure-Color Scenes
浏览论文内容
中文总结 AI 辅助
针对纯彩色场景中色度线索坍缩导致色彩恒常性估计歧义的问题，提出结合SAFE网络与LCS的框架，显著降低了平均及高低分位数的角度误差。
中文摘要 AI 辅助
纯彩色场景下的色彩恒常性极具挑战性：当大部分像素共享狭窄的色调范围时，所有基于色度的线索都会坍缩为单个点，导致标准估计器出现歧义。我们提出了一个紧凑框架，结合两项创新：（i）SAFE，即场景感知特征调制网络，它将光照线索组织为结构化的四令牌表示，随后基于场景复杂度特征对其进行选择性重加权；（ii）学习色彩空间（Learned Color Space，LCS），这是一种依赖于场景的色度归一化方法，直接解决纯彩色场景中的色度坍缩问题。实验结果表明，SAFE在纯彩色场景中持续提升性能：与各指标中表现最佳的基线相比，它将平均角度误差降低了10%，最佳25%误差降低了20%，最差25%误差降低了5.8%。
英文摘要
Color constancy on pure-color scenes is challenging: when most pixels share a narrow band of hues, every chromaticity-based cue collapses to a single point and standard estimators become ambiguous. We propose a compact framework that couples two innovations: (i) SAFE, a Scene-Aware FeaturE modulation network that organizes illumination cues into a structured four-token representation, which is then selectively reweighted based on scene complexity features; (ii) the Learned Color Space (LCS), a scene-dependent chromaticity normalization that directly addresses the chromaticity collapse problem for pure-color scenes. Experiment results show that SAFE consistently improves performance in pure-color scenes. Compared to the best-performing baseline in each metric, it reduces the mean angular error by 10%, the best-25% error by 20%, and the worst-25% error by 5.8%.
发表机构
- MediaTek Inc.（联发科技股份有限公司）
- National Taiwan University（台湾大学）
机构由 AI 辅助整理，请以论文原文为准。
