Light 'em Up: Enabling Few-Shot Low-Light 3D Gaussian Splatting with Multi-Scale Explicit Retinex Illumination Decoupling

TL;DR

MERID-GS is a novel framework that enables high-quality, few-shot, low-light 3D view synthesis by explicitly decoupling illumination and reflectance using Retinex theory, improving generalization and stability.

Contribution

It introduces MERID-GS, a multi-scale explicit Retinex-based approach combined with 3D Gaussian Splatting for robust low-light 3D synthesis from sparse views.

Findings

Achieves state-of-the-art performance in low-light 3D synthesis.

Demonstrates superior cross-scene generalization and view consistency.

Enables stable synthesis with only a few shots.

Abstract

Full 360$^\circ$ novel view synthesis under low-light conditions remains challenging. Insufficient illumination, noise amplification, and view-dependent photometric inconsistencies prevent existing methods from jointly preserving geometric consistency and photorealism. Unsupervised approaches often exhibit color drift under large viewpoint variations, while supervised low-light enhancement models, though effective for 2D tasks, struggle to generalize to new scenes and typically require retraining. To address this issue, we propose MERID-GS, a Multi-Scale Explicit Retinex Illumination-Decoupled Gaussian framework for low-light 360$^\circ$ synthesis. Based on Retinex theory, the method explicitly separates illumination and reflectance, and suppresses noise propagation while enhancing dark-region structures via a learnable gain and Illumination-State-Guided Frequency Gating. Combined with lightweight Reflection Head and 3D Gaussian Splatting, MERID-GS adapts to new scenes with only a few shots and enables stable low-light novel view synthesis from sparse-view observations. In addition, we construct a low-light multi-view dataset covering full 360$^\circ$ scenes for joint evaluation. Thorough experiments across multiple datasets in this area demonstrate that MERID-GS achieves SOTA performance, exhibiting superior cross-scene generalization and view consistency. The source code and pre-trained models are available at https://github.com/YhuoyuH/MERID-GS..