arXiv 2604.033092026-03-31cs.CVcs.AI

TreeGaussian：基于树引导的级联对比学习用于层次一致的3D高斯场景分割与理解

TreeGaussian: Tree-Guided Cascaded Contrastive Learning for Hierarchical Consistent 3D Gaussian Scene Segmentation and Understanding

Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Beihang University（北京航空航天大学）

Jingbin You, Zehao Li, Hao Jiang, Xinzhu Ma, Shuqin Gao, Honglong Zhao, Congcong Zheng, Tianlu Mao, Feng Dai, Yucheng Zhang, Zhaoqi Wang

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出TreeGaussian，通过树引导的级联对比学习框架，解决3DGS在表示层次结构和捕捉整体-部分关系上的不足，提升分割一致性与质量。

摘要：

3D高斯溅射（3DGS）已发展为一种实时、可微分的神经场景理解表示。然而，现有基于3DGS的方法难以表示层次化的3D语义结构并捕捉复杂场景中的整体-部分关系。此外，密集的成对比较和不一致的层次标签会阻碍特征学习，导致分割效果不佳。为解决这些问题，我们引入TreeGaussian，一种树引导的级联对比学习框架，明确建模层次语义关系并减少对比监督的冗余。通过构建多级对象树，TreeGaussian实现了跨对象-部分层次的结构化学习。此外，我们提出了一种两阶段级联对比学习策略，逐步从全局到局部优化特征表示，缓解饱和问题并稳定训练。一致分割检测（CSD）机制和基于图的去噪模块进一步用于对齐不同视角的分割模式并抑制不稳定的高斯点，提升分割的一致性和质量。广泛的实验，包括开放词汇3D物体选择、3D点云理解和消融研究，证明了本方法的有效性和鲁棒性。

英文摘要：

3D Gaussian Splatting (3DGS) has emerged as a real-time, differentiable representation for neural scene understanding. However, existing 3DGS-based methods struggle to represent hierarchical 3D semantic structures and capture whole-part relationships in complex scenes. Moreover, dense pairwise comparisons and inconsistent hierarchical labels from 2D priors hinder feature learning, resulting in suboptimal segmentation. To address these limitations, we introduce TreeGaussian, a tree-guided cascaded contrastive learning framework that explicitly models hierarchical semantic relationships and reduces redundancy in contrastive supervision. By constructing a multi-level object tree, TreeGaussian enables structured learning across object-part hierarchies. In addition, we propose a two-stage cascaded contrastive learning strategy that progressively refines feature representations from global to local, mitigating saturation and stabilizing training. A Consistent Segmentation Detection (CSD) mechanism and a graph-based denoising module are further introduced to align segmentation modes across views while suppressing unstable Gaussian points, enhancing segmentation consistency and quality. Extensive experiments, including open-vocabulary 3D object selection, 3D point cloud understanding, and ablation studies, demonstrate the effectiveness and robustness of our approach.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。