Semantic segmentation on point clouds is critical for 3D scene understanding. However, sparse and irregular point distributions provide limited appearance evidence, making geometry-only features insufficient to distinguish objects with similar shapes but distinct appearances (e.g., color, texture, material). We propose Gaussian-to-Point (G2P), which transfers appearance-aware attributes from 3D Gaussian Splatting to point clouds for more discriminative and appearance-consistent segmentation. Our G2P address the misalignment between optimized Gaussians and original point geometry by establishing point-wise correspondences. By leveraging Gaussian opacity attributes, we resolve the geometric ambiguity that limits existing models. Additionally, Gaussian scale attributes enable precise boundary localization in complex 3D scenes. Extensive experiments demonstrate that our approach achieves superior performance on standard benchmarks and shows significant improvements on geometrically challenging classes, all without any 2D or language supervision.
@article{arxiv.2601.03510,
title = {G2P: Gaussian-to-Point Attribute Alignment for Boundary-Aware 3D Semantic Segmentation},
author = {Hojun Song and Chae-yeong Song and Jeong-hun Hong and Chaewon Moon and Dong-hwi Kim and Gahyeon Kim and Soo Ye Kim and Yiyi Liao and Jaehyup Lee and Sang-hyo Park},
journal= {arXiv preprint arXiv:2601.03510},
year = {2026}
}