Summary
TokenLight (arXiv:2504.13097) is an image relighting method by Sumit Chaturvedi, Yannick Hold-Geoffroy, and Mengwei Ren that provides precise, continuous control over multiple illumination attributes in a photograph. The authors formulate relighting as a conditional image generation task and introduce attribute tokens that encode distinct lighting factors such as intensity, color, ambient illumination, diffuse level, and 3D light positions. The model is trained on a large-scale synthetic dataset with ground-truth lighting annotations, supplemented by a small set of real captures to improve realism and generalization. It was validated on a variety of relighting tasks, including controlling in-scene lighting fixtures and editing environment illumination with virtual light sources, on both synthetic and real images. Compared with prior work, TokenLight achieves state-of-the-art quantitative and qualitative results. Notably, without explicit inverse-rendering supervision, the model demonstrates an implicit understanding of how light interacts with scene geometry, occlusion, and materials, producing convincing effects even in challenging scenarios such as placing light sources inside objects or relighting transparent materials. Project page: vrroom.github.io/tokenlight/.
Paper: TokenLight: Precise Lighting Control in Images using Attribute Tokens
Authors: Sumit Chaturvedi, Yannick Hold-Geoffroy, Mengwei Ren
Published: 2025-04-17
Field: Computer Vision
Overview
This paper presents a method for image relighting that enables precise and continuous control over multiple illumination attributes in a photograph. The authors formulate relighting as a conditional image generation task and introduce attribute tokens to encode distinct lighting factors such as intensity, color, ambient illumination, diffuse level, and 3D light positions.
Key Points
- Formulates relighting as conditional image generation with attribute tokens encoding individual lighting factors.
- Trained on a large-scale synthetic dataset with ground-truth lighting annotations, supplemented by a small set of real captures for realism and generalization.
- Validated across tasks including controlling in-scene lighting fixtures and editing environment illumination using virtual light sources, on both synthetic and real images.
- Achieves state-of-the-art quantitative and qualitative performance compared to prior relighting methods.
- Without explicit inverse-rendering supervision, the model shows an implicit understanding of how light interacts with scene geometry, occlusion, and materials — producing convincing results even in traditionally challenging scenarios, such as placing a light source inside an object or relighting transparent materials.
Project page:
vrroom.github.io/tokenlight/
This page is an English static mirror generated for search and AI citation.
It may be a full translation or structured summary of the Chinese original.
Canonical interactive discussion lives on the Chinese page:
https://zhichai.net/topic/177618537