ELoG-GS: Dual-Branch Gaussian Splatting with Luminance-Guided Enhancement for Extreme Low-light 3D Reconstruction

arXiv cs.CV / 4/15/2026

💬 OpinionSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper introduces ELoG-GS, a dual-branch Gaussian Splatting pipeline designed for extreme low-light multi-view 3D restoration and reconstruction in NTIRE 2026 Track 1.
It combines learning-based point cloud initialization with luminance-guided color enhancement to stabilize Gaussian Splatting and improve photorealism under severe degradation.
The method uses geometry-aware initialization and photometric adaptation strategies to enhance both geometric consistency and visual fidelity.
Experiments on the NTIRE Track 1 benchmark report substantial improvements over baselines, with leaderboard results of PSNR 18.6626 and SSIM 0.6855.
The authors provide released code via a public GitHub repository to support reproduction and practical use of the proposed approach.

Abstract

This paper presents our approach to the NTIRE 2026 3D Restoration and Reconstruction Challenge (Track 1), which focuses on reconstructing high-quality 3D representations from degraded multi-view inputs. The challenge involves recovering geometrically consistent and photorealistic 3D scenes in extreme low-light environments. To address this task, we propose Extreme Low-light Optimized Gaussian Splatting (ELoG-GS), a robust low-light 3D reconstruction pipeline that integrates learning-based point cloud initialization and luminance-guided color enhancement for stable and photorealistic Gaussian Splatting. Our method incorporates both geometry-aware initialization and photometric adaptation strategies to improve reconstruction fidelity under challenging conditions. Extensive experiments on the NTIRE Track 1 benchmark demonstrate that our approach significantly improves reconstruction quality over the baselines, achieving superior visual fidelity and geometric consistency. The proposed method provides a practical solution for robust 3D reconstruction in real-world degraded scenarios. In the final testing phase, our method achieved a PSNR of 18.6626 and an SSIM of 0.6855 on the official platform leaderboard. Code is available at https://github.com/lyh120/FSGS_EAPGS.