After the Party: Navigating the Mapping From Color to Ambient Lighting

2025-08-04

AI Summary

Existing image restoration methods typically assume uniform or single-source white-balanced illumination, rendering them inadequate for complex scenes with heterogeneous lighting—such as multiple colored light sources, occlusions, and spatially varying material reflectance—leading to illumination inconsistency, texture leakage, and chromatic distortion. To address this, we introduce CL3AN, the first high-resolution dataset specifically designed for complex colored illumination. Building upon Retinex theory, we propose a novel dual-guided learning framework that explicitly decouples illumination and reflectance by jointly modeling chromaticity and luminance. Unlike prior approaches, our method requires no strong priors and achieves state-of-the-art performance across multiple benchmarks. It demonstrates superior robustness, significantly reduced computational overhead, and effectively mitigates both color distortion and texture leakage.

Technology Category

• Computer Vision: Computational Photography, Image & Video Synthesis
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Illumination in practical scenarios is inherently complex, involving colored light sources, occlusions, and diverse material interactions that produce intricate reflectance and shading effects. However, existing methods often oversimplify this challenge by assuming a single light source or uniform, white-balanced lighting, leaving many of these complexities unaddressed.In this paper, we introduce CL3AN, the first large-scale, high-resolution dataset of its kind designed to facilitate the restoration of images captured under multiple Colored Light sources to their Ambient-Normalized counterparts. Through benchmarking, we find that leading approaches often produce artifacts, such as illumination inconsistencies, texture leakage, and color distortion, primarily due to their limited ability to precisely disentangle illumination from reflectance. Motivated by this insight, we achieve such a desired decomposition through a novel learning framework that leverages explicit chromaticity and luminance components guidance, drawing inspiration from the principles of the Retinex model. Extensive evaluations on existing benchmarks and our dataset demonstrate the effectiveness of our approach, showcasing enhanced robustness under non-homogeneous color lighting and material-specific reflectance variations, all while maintaining a highly competitive computational cost. The benchmark, codes, and models are available at www.github.com/fvasluianu97/RLN2.

Problem

• Restoring images under multiple colored light sources
• Disentangling illumination from reflectance accurately
• Addressing artifacts like color distortion and texture leakage

Innovation

• Large-scale dataset for colored light restoration
• Chromaticity and luminance guided learning framework
• Robust decomposition under non-homogeneous lighting