NG-GS: NeRF-Guided 3D Gaussian Splatting Segmentation

2026-04-16

AI Summary

This work addresses the boundary aliasing and segmentation artifacts in 3D Gaussian Splatting (3DGS) caused by its discrete representation. To mitigate these issues, the paper introduces— for the first time—a NeRF-guided continuous feature field for 3DGS segmentation. The approach identifies ambiguous Gaussians near object boundaries using mask variance, constructs a continuous semantic field via radial basis function (RBF) interpolation, and jointly optimizes this field with a lightweight NeRF module enhanced by multi-resolution hash encoding. A spatial continuity loss further regularizes the optimization. This integrated framework substantially alleviates boundary discretization artifacts and achieves state-of-the-art performance on NVOS, LERF-OVS, and ScanNet benchmarks, with notably improved boundary mIoU.

Technology Category

• Computer Vision: Segmentation
• Search and Optimization: Mixed Discrete/Continuous Search
• Machine Learning: Neuro-Symbolic Learning

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

Recent advances in 3D Gaussian Splatting (3DGS) have enabled highly efficient and photorealistic novel view synthesis. However, segmenting objects accurately in 3DGS remains challenging due to the discrete nature of Gaussian representations, which often leads to aliasing and artifacts at object boundaries. In this paper, we introduce NG-GS, a novel framework for high-quality object segmentation in 3DGS that explicitly addresses boundary discretization. Our approach begins by automatically identifying ambiguous Gaussians at object boundaries using mask variance analysis. We then apply radial basis function (RBF) interpolation to construct a spatially continuous feature field, enhanced by multi-resolution hash encoding for efficient multi-scale representation. A joint optimization strategy aligns 3DGS with a lightweight NeRF module through alignment and spatial continuity losses, ensuring smooth and consistent segmentation boundaries. Extensive experiments on NVOS, LERF-OVS, and ScanNet benchmarks demonstrate that our method achieves state-of-the-art performance, with significant gains in boundary mIoU. Code is available at https://github.com/BJTU-KD3D/NG-GS.

Problem

• 3D Gaussian Splatting
• object segmentation
• boundary discretization
• aliasing
• photorealistic novel view synthesis

Innovation

• 3D Gaussian Splatting
• object segmentation
• NeRF-guided
• RBF interpolation
• multi-resolution hash encoding