GLOW: Global Illumination-Aware Inverse Rendering of Indoor Scenes Captured with Dynamic Co-Located Light & Camera

2025-11-27

AI Summary

Indoor inverse rendering suffers from inherent ambiguity due to strong coupling between reflectance and illumination—especially under complex interreflections among multiple objects. While natural-light approaches struggle to disentangle these factors, co-located camera-light setups enable illumination calibration but introduce new challenges: dynamic shadows, near-field lighting effects, strong interreflections, and motion-induced specular highlights. This paper proposes a global illumination-aware neural implicit representation framework. We introduce a dynamic radiance cache and a surface-angle-weighted radiance loss to explicitly model illumination-induced discontinuities and specular artifacts. By jointly optimizing neural implicit geometry, differentiable rendering, structure-from-motion (SfM) calibration, and reflectance estimation, our method achieves coupled geometric and albedo reconstruction. Extensive experiments demonstrate state-of-the-art albedo estimation accuracy under both natural and co-located lighting conditions, validating robust modeling of complex indoor illumination.

Technology Category

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

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Inverse rendering of indoor scenes remains challenging due to the ambiguity between reflectance and lighting, exacerbated by inter-reflections among multiple objects. While natural illumination-based methods struggle to resolve this ambiguity, co-located light-camera setups offer better disentanglement as lighting can be easily calibrated via Structure-from-Motion. However, such setups introduce additional complexities like strong inter-reflections, dynamic shadows, near-field lighting, and moving specular highlights, which existing approaches fail to handle. We present GLOW, a Global Illumination-aware Inverse Rendering framework designed to address these challenges. GLOW integrates a neural implicit surface representation with a neural radiance cache to approximate global illumination, jointly optimizing geometry and reflectance through carefully designed regularization and initialization. We then introduce a dynamic radiance cache that adapts to sharp lighting discontinuities from near-field motion, and a surface-angle-weighted radiometric loss to suppress specular artifacts common in flashlight captures. Experiments show that GLOW substantially outperforms prior methods in material reflectance estimation under both natural and co-located illumination.

Problem

• Resolves ambiguity between reflectance and lighting in indoor scenes
• Handles strong inter-reflections, dynamic shadows, and moving specular highlights
• Improves material reflectance estimation under co-located illumination setups

Innovation

• Neural implicit surface with radiance cache for global illumination
• Dynamic radiance cache adapts to near-field lighting discontinuities
• Surface-angle-weighted loss suppresses specular artifacts in captures