A full-body 3D avatar model that combines learned zonal-harmonic light transport, a shadow network, and deferred shading for relighting and animation of body, face, and hands.
SwitchLight: Co-design of Physics-driven Architecture and Pre-training Framework for Human Portrait Relighting
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
We introduce a co-designed approach for human portrait relighting that combines a physics-guided architecture with a pre-training framework. Drawing on the Cook-Torrance reflectance model, we have meticulously configured the architecture design to precisely simulate light-surface interactions. Furthermore, to overcome the limitation of scarce high-quality lightstage data, we have developed a self-supervised pre-training strategy. This novel combination of accurate physical modeling and expanded training dataset establishes a new benchmark in relighting realism.
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Relightable Full-Body Gaussian Codec Avatars
A full-body 3D avatar model that combines learned zonal-harmonic light transport, a shadow network, and deferred shading for relighting and animation of body, face, and hands.