ClunoGS: Compact Language 3D Gaussian Splatting via Unified Optimization
作者: Jian Xu, Yi Li, Ying Huang, Siyi Wu, Bingzhe Li, Yuede Ji, Miao Yin
分类: cs.CV
发布日期: 2026-10-21
💡 一句话要点
提出ClunoGS以解决语言3D高斯点云压缩问题
🎯 匹配领域: 支柱三:空间感知与语义 (Perception & Semantics)
关键词: 3D高斯点云 语义理解 压缩方法 向量量化 稀疏化 开放词汇 深度学习
📋 核心要点
- 现有的3D高斯点云压缩方法主要关注RGB渲染质量,导致重要的语义信息被丢失。
- ClunoGS通过结合语义感知的高斯稀疏化和向量量化,提出了一种统一的优化框架。
- 在LERF和ScanNet-10基准测试中,ClunoGS在激进压缩下显著提高了语义表示的保留效果。
📝 摘要(中文)
语言3D高斯点云(3D Gaussian Splatting)通过将高斯原语与语言特征关联,实现开放词汇场景理解。然而,现有的3DGS压缩方法主要关注RGB渲染质量,导致语义信息的丢失。为此,本文提出ClunoGS,一个紧凑的语言3DGS框架,结合了语义感知的高斯稀疏化和向量量化,采用统一的约束目标进行优化。通过在稀疏化过程中引入语义监督,ClunoGS在渲染质量、语义准确性和紧凑性之间进行联合优化。实验结果表明,该方法在开放词汇定位和分割任务中显著减轻了因序列压缩和无语义感知稀疏化造成的语义表示退化。
🔬 方法详解
问题定义:本文旨在解决现有3D高斯点云压缩方法在保留语义信息方面的不足,尤其是在仅基于渲染质量进行稀疏化时,可能会去除对语义预测重要的高斯原语。
核心思路:ClunoGS的核心思路是将语义感知的高斯稀疏化与向量量化结合在一个统一的优化框架中,通过引入语义监督来提升语义准确性。
技术框架:该框架包括四个主要模块:稀疏化、量化、原始更新和对偶更新。通过交替优化这些模块,ClunoGS能够在渲染质量、语义准确性和紧凑性之间达到平衡。
关键创新:ClunoGS的创新在于其统一的约束目标,能够同时考虑渲染质量和语义信息的保留,避免了传统方法中语义信息的丢失。
关键设计:在设计中,采用了语义监督作为稀疏化过程中的指导,损失函数同时考虑了渲染质量和语义准确性,确保了高斯原语的有效保留。
🖼️ 关键图片
📊 实验亮点
在LERF和ScanNet-10基准测试中,ClunoGS在激进压缩条件下,语义表示的保留效果显著优于传统方法,具体表现为在语义准确性上提升了XX%,有效减轻了信息丢失问题。
🎯 应用场景
ClunoGS的研究成果具有广泛的应用潜力,特别是在自动驾驶、机器人导航和增强现实等领域。通过提高语义理解能力,该方法能够帮助系统更好地理解和处理复杂场景,从而提升决策和交互的智能化水平。
📄 摘要(原文)
Language 3D Gaussian Splatting enables open-vocabulary scene understanding by associating Gaussian primitives with language features. These semantic attributes increase model storage, while most existing 3DGS compression methods are designed mainly to preserve RGB rendering quality. One straightforward approach is to apply sparsification and quantization sequentially. However, pruning based only on rendering quality can remove Gaussians that have small RGB contributions but are important for semantic prediction. Once removed, their semantic information cannot be recovered by later quantization. We propose ClunoGS, a compact framework for Language 3DGS that combines semantics-aware Gaussian sparsification and vector quantization in a unified constrained objective. Semantic supervision is used during sparsification, and the objective jointly considers rendering quality, semantic accuracy, and compactness. We optimize the unified objective with alternating primal, sparsification, quantization, and dual updates. We evaluate ClunoGS on open-vocabulary localization and segmentation using the LERF and ScanNet-10 benchmarks. Under aggressive compression, our unified, semantics-aware formulation substantially alleviates the degradation of semantic representations caused by sequential compression and semantics-agnostic Gaussian sparsification. Ablations further show that jointly coordinating sparsification and quantization better preserves semantic accuracy than applying them independently.