English

SAGD: Boundary-Enhanced Segment Anything in 3D Gaussian via Gaussian Decomposition

Computer Vision and Pattern Recognition 2025-01-22 v4

Abstract

3D Gaussian Splatting has emerged as an alternative 3D representation for novel view synthesis, benefiting from its high-quality rendering results and real-time rendering speed. However, the 3D Gaussians learned by 3D-GS have ambiguous structures without any geometry constraints. This inherent issue in 3D-GS leads to a rough boundary when segmenting individual objects. To remedy these problems, we propose SAGD, a conceptually simple yet effective boundary-enhanced segmentation pipeline for 3D-GS to improve segmentation accuracy while preserving segmentation speed. Specifically, we introduce a Gaussian Decomposition scheme, which ingeniously utilizes the special structure of 3D Gaussian, finds out, and then decomposes the boundary Gaussians. Moreover, to achieve fast interactive 3D segmentation, we introduce a novel training-free pipeline by lifting a 2D foundation model to 3D-GS. Extensive experiments demonstrate that our approach achieves high-quality 3D segmentation without rough boundary issues, which can be easily applied to other scene editing tasks.

Keywords

Cite

@article{arxiv.2401.17857,
  title  = {SAGD: Boundary-Enhanced Segment Anything in 3D Gaussian via Gaussian Decomposition},
  author = {Xu Hu and Yuxi Wang and Lue Fan and Chuanchen Luo and Junsong Fan and Zhen Lei and Qing Li and Junran Peng and Zhaoxiang Zhang},
  journal= {arXiv preprint arXiv:2401.17857},
  year   = {2025}
}