Naka-GS: A Bionics-inspired Dual-Branch Naka Correction and Progressive Point Pruning for Low-Light 3DGS

2026-04-13

AI Summary

This work addresses the degraded reconstruction quality of 3D Gaussian splatting under low-light conditions, where poor visibility, color distortion, and corrupted geometric priors pose significant challenges. To tackle these issues, the authors propose a bio-inspired dual-branch framework that jointly optimizes photometric restoration and geometric initialization by integrating physics-informed priors with frequency-domain decoupled correction. Additionally, a lightweight point preprocessing module is introduced, which performs structure-preserving outlier removal through mask-guided filtering, coordinate alignment, voxel pooling, and distance-adaptive progressive pruning. Evaluated on the NTIRE 3DRR Challenge, the proposed method substantially outperforms baseline approaches, achieving consistent improvements in reconstruction fidelity, training stability, and optimization efficiency.

Technology Category

• Computer Vision: 3D Computer Vision
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Low-light conditions severely hinder 3D restoration and reconstruction by degrading image visibility, introducing color distortions, and contaminating geometric priors for downstream optimization. We present NAKA-GS, a bionics-inspired framework for low-light 3D Gaussian Splatting that jointly improves photometric restoration and geometric initialization. Our method starts with a Naka-guided chroma-correction network, which combines physics-prior low-light enhancement, dual-branch input modeling, frequency-decoupled correction, and mask-guided optimization to suppress bright-region chromatic artifacts and edge-structure errors. The enhanced images are then fed into a feed-forward multi-view reconstruction model to produce dense scene priors. To further improve Gaussian initialization, we introduce a lightweight Point Preprocessing Module (PPM) that performs coordinate alignment, voxel pooling, and distance-adaptive progressive pruning to remove noisy and redundant points while preserving representative structures. Without introducing heavy inference overhead, NAKA-GS improves restoration quality, training stability, and optimization efficiency for low-light 3D reconstruction. The proposed method was presented in the NTIRE 3D Restoration and Reconstruction (3DRR) Challenge, and outperformed the baseline methods by a large margin. The code is available at https://github.com/RunyuZhu/Naka-GS

Problem

• low-light
• 3D reconstruction
• image restoration
• geometric priors
• color distortion

Innovation

• Naka correction
• dual-branch modeling
• progressive point pruning
• low-light 3D Gaussian Splatting
• bionics-inspired