GaussianTrimmer: Online Trimming Boundaries for 3DGS Segmentation

2026-01-19

AI Summary

Existing 3D Gaussian-based segmentation methods often produce jagged boundaries at foreground–background interfaces due to significant scale variations among Gaussians. To address this issue, this work proposes a plug-and-play online boundary refinement mechanism that leverages uniformly distributed virtual cameras to capture multi-view 2D semantic segmentations, which are then fused to prune boundaries directly at the level of the original 3D Gaussian primitives. This approach is the first to employ virtual-view-guided 2D segmentation for optimizing 3D Gaussian boundaries, effectively mitigating boundary blurring caused by large-scale Gaussians spanning multiple semantic regions. Experimental results demonstrate that the proposed framework consistently enhances boundary accuracy across various 3D Gaussian segmentation methods, offering both strong generality and effectiveness.

Technology Category

• Computer Vision: Segmentation
• Search and Optimization: Mixed Discrete/Continuous Search
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Virtualization and resource management in Web systems and infrastructures
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

With the widespread application of 3D Gaussians in 3D scene representation, 3D scene segmentation methods based on 3D Gaussians have also gradually emerged. However, existing 3D Gaussian segmentation methods basically segment on the basis of Gaussian primitives. Due to the large variation range of the scale of 3D Gaussians, large-sized Gaussians that often span the foreground and background lead to jagged boundaries of segmented objects. To this end, we propose an online boundary trimming method, GaussianTrimmer, which is an efficient and plug-and-play post-processing method capable of trimming coarse boundaries for existing 3D Gaussian segmentation methods. Our method consists of two core steps: 1. Generating uniformly and well-covered virtual cameras; 2. Trimming Gaussian at the primitive level based on 2D segmentation results on virtual cameras. Extensive quantitative and qualitative experiments demonstrate that our method can improve the segmentation quality of existing 3D Gaussian segmentation methods as a plug-and-play method.

Problem

• 3D Gaussian Splatting
• scene segmentation
• boundary artifacts
• object boundaries
• segmentation quality

Innovation

• 3D Gaussian Splatting
• scene segmentation
• boundary refinement
• virtual camera
• plug-and-play post-processing