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

Jiaye Wu1, Saeed Hadadan1, Geng Lin1, Peihan Tu1, Matthias Zwicker1, David Jacobs1, Roni Sengupta2
1University of Maryland, College Park 2University of North Carolina, Chapel Hill
CVPRF 2026
Teaser for GLOW showing inverse rendering results, geometry, and material properties.

GLOW reconstructs indoor scenes from images captured with a co-located light and camera, recovering geometry and material properties while accounting for challenging global illumination.

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.

Experimental Results

Comparison of our technique against natural-illumination and co-located light and camera baselines on synthetic data.

Bedroom

Natural illumination
Co-located light & camera
InputBedroom synthetic input rendering under natural illumination
NeROBedroom albedo estimated by NeRO
IRGSBedroom albedo estimated by IRGS
InputBedroom synthetic input rendering under co-located light and camera
IRONBedroom albedo estimated by IRON
WildLightBedroom albedo estimated by WildLight
OursBedroom albedo estimated by Ours
GTBedroom ground-truth albedo

Shelf

Natural illumination
Co-located light & camera
InputShelf synthetic input rendering under natural illumination
NeROShelf albedo estimated by NeRO
IRGSShelf albedo estimated by IRGS
InputShelf synthetic input rendering under co-located light and camera
IRONShelf albedo estimated by IRON
WildLightShelf albedo estimated by WildLight
OursShelf albedo estimated by Ours
GTShelf ground-truth albedo

Kitchen

Natural illumination
Co-located light & camera
InputKitchen synthetic input rendering under natural illumination
NeROKitchen albedo estimated by NeRO
IRGSKitchen albedo estimated by IRGS
InputKitchen synthetic input rendering under co-located light and camera
IRONKitchen albedo estimated by IRON
WildLightKitchen albedo estimated by WildLight
OursKitchen albedo estimated by Ours
GTKitchen ground-truth albedo

Supplementary Videos

Rerendering
Albedo
Roughness
Normal

Poster

BibTeX

@article{wu2026glow,
  title={GLOW: Global Illumination-Aware Inverse Rendering of Indoor Scenes Captured with Dynamic Co-Located Light and Camera},
  author={Wu, Jiaye and Hadadan, Saeed and Lin, Geng and Tu, Peihan and Zwicker, Matthias and Jacobs, David and Sengupta, Roni},
  journal={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year={2026}
}