High Dynamic Range 3D Gaussian Splatting via Luminance-Chromaticity Decomposition

TL;DR

This paper introduces LCD-GS, a novel HDR 3D Gaussian Splatting method that decomposes luminance and chromaticity, improving reconstruction quality and enabling luminance editing with minimal model complexity.

Contribution

Proposes luminance-chromaticity decomposition for HDR 3D Gaussian Splatting, enhancing learning flexibility and performance without complex architectures.

Findings

LCD-GS outperforms state-of-the-art methods in fidelity and dynamic range.

Decoupling luminance and chromaticity improves HDR 3D reconstruction.

Minimal parameter increase allows primitive-level luminance editing.

Abstract

High Dynamic Range (HDR) 3D reconstruction is pivotal for professional content creation in filmmaking and virtual production. Existing methods typically rely on multi-exposure Low Dynamic Range (LDR) supervision to constrain the learning process within vast brightness spaces, resulting in complex, dual-branch architectures. This work explores the feasibility of learning HDR 3D models exclusively in the HDR data space to simplify model design. By analyzing 3D Gaussian Splatting (3DGS) for HDR imagery, we reveal that its failure stems from the limited capacity of Spherical Harmonics (SHs) to capture extreme radiance variations across views, often biasing towards high-radiance observations and underfitting. While increasing the maximum SH degree improves training fitting, it leads to severe overfitting and excessive parameter overhead. To address this, we propose \textit{Luminance--Chromaticity Decomposition Gaussian Splatting} (LCD-GS). By decoupling luminance and chromaticity into independent parameters, LCD-GS significantly enhances learning flexibility with minimal parameter increase (\textit{e.g.}, one extra scalar per primitive). Notably, LCD-GS maintains the original training and inference pipeline, requiring only a change in color representation. This explicit decomposition naturally enables primitive-level local and global luminance editing during inference. Extensive experiments on synthetic and real datasets demonstrate that LCD-GS consistently outperforms state-of-the-art methods in reconstruction fidelity and dynamic-range preservation even with a simpler, more efficient architecture, providing an elegant paradigm for professional-grade HDR 3D modeling. Code and datasets will be released.