Clean-GS: Semantic Mask-Guided Pruning for 3D Gaussian Splatting

TL;DR

Clean-GS is a novel semantic mask-guided pruning method that significantly reduces 3D Gaussian Splatting model sizes by removing artifacts and background clutter, enabling practical deployment in bandwidth-limited scenarios.

Contribution

It introduces a multi-stage pruning approach using minimal semantic masks to effectively remove irrelevant Gaussians from 3D reconstructions, outperforming existing global importance-based methods.

Findings

Achieves 60-80% model compression while preserving quality

Reduces file size from 125MB to 47MB on benchmark datasets

Enables practical web and AR/VR deployment of 3DGS models

Abstract

3D Gaussian Splatting produces high-quality scene reconstructions but generates hundreds of thousands of spurious Gaussians (floaters) scattered throughout the environment. These artifacts obscure objects of interest and inflate model sizes, hindering deployment in bandwidth-constrained applications. We present Clean-GS, a method for removing background clutter and floaters from 3DGS reconstructions using sparse semantic masks. Our approach combines whitelist-based spatial filtering with color-guided validation and outlier removal to achieve 60-80\% model compression while preserving object quality. Unlike existing 3DGS pruning methods that rely on global importance metrics, Clean-GS uses semantic information from as few as 3 segmentation masks (1\% of views) to identify and remove Gaussians not belonging to the target object. Our multi-stage approach consisting of (1) whitelist filtering via projection to masked regions, (2) depth-buffered color validation, and (3) neighbor-based outlier removal isolates monuments and objects from complex outdoor scenes. Experiments on Tanks and Temples show that Clean-GS reduces file sizes from 125MB to 47MB while maintaining rendering quality, making 3DGS models practical for web deployment and AR/VR applications. Our code is available at https://github.com/smlab-niser/clean-gs