HeadLighter: Disentangling Illumination in Generative 3D Gaussian Heads via Lightstage Captures

2026-01-05

arXiv.org

AI Summary

Existing 3D Gaussian splatting-based head generation methods struggle to disentangle illumination from intrinsic appearance, limiting controllable relighting capabilities. This work proposes a supervised dual-branch architecture that, for the first time, leverages multi-view images under controlled lighting—captured by a light-stage system—as supervision signals. Through progressive disentanglement training and normal distillation, the method effectively separates illumination-invariant intrinsic appearance from physically plausible rendering components. The approach achieves high-fidelity, real-time rendering of editable 3D heads while enabling explicit control over both lighting and viewpoint, and demonstrates accurate relighting performance even under complex illumination conditions.

Technology Category

• Computer Vision: 3D Computer Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Recent 3D-aware head generative models based on 3D Gaussian Splatting achieve real-time, photorealistic and view-consistent head synthesis. However, a fundamental limitation persists: the deep entanglement of illumination and intrinsic appearance prevents controllable relighting. Existing disentanglement methods rely on strong assumptions to enable weakly supervised learning, which restricts their capacity for complex illumination. To address this challenge, we introduce HeadLighter, a novel supervised framework that learns a physically plausible decomposition of appearance and illumination in head generative models. Specifically, we design a dual-branch architecture that separately models lighting-invariant head attributes and physically grounded rendering components. A progressive disentanglement training is employed to gradually inject head appearance priors into the generative architecture, supervised by multi-view images captured under controlled light conditions with a light stage setup. We further introduce a distillation strategy to generate high-quality normals for realistic rendering. Experiments demonstrate that our method preserves high-quality generation and real-time rendering, while simultaneously supporting explicit lighting and viewpoint editing. We will publicly release our code and dataset.

Problem

• illumination disentanglement
• 3D head generation
• relighting
• appearance-illumination entanglement
• 3D Gaussian Splatting

Innovation

• 3D Gaussian Splatting
• illumination disentanglement
• light stage
• relighting
• generative head modeling