LESV: 语言嵌入稀疏体素融合用于开放词汇3D场景理解
LESV: Language Embedded Sparse Voxel Fusion for Open-Vocabulary 3D Scene Understanding
AI总结：
本文提出LESV框架，通过稀疏体素 rasterization 解决3D场景理解中空间和语义模糊问题，利用AM-RADIO实现密集对齐，提升细粒度查询性能。
摘要：
近期开放词汇3D场景理解的进步严重依赖于3D高斯散射（3DGS）将视觉-语言特征注册到3D空间。然而，我们发现这些方法存在两个关键限制：无结构重叠高斯导致的空间模糊，需要概率特征注册；以及通过物体级掩码池化导致的多级语义模糊，稀释了细粒度细节。为解决这些挑战，我们提出一个新框架，利用稀疏体素 rasterization（SVRaster）作为结构化、不重叠的几何表示。通过用单目深度和法线先验正则化SVRaster，我们建立了稳定的几何基础。这使特征注册过程确定性且具有置信度意识，并抑制了3DGS中常见的语义出血伪影。此外，我们通过利用新兴的密集对齐性质的foundation model AM-RADIO解决多级模糊问题，避免了分层训练方法的计算开销。我们的方法在开放词汇3D物体检索和点云理解基准上实现了最先进的性能，特别是在细粒度查询中表现突出，而注册方法通常失败。
英文摘要：
Recent advancements in open-vocabulary 3D scene understanding heavily rely on 3D Gaussian Splatting (3DGS) to register vision-language features into 3D space. However, we identify two critical limitations in these approaches: the spatial ambiguity arising from unstructured, overlapping Gaussians which necessitates probabilistic feature registration, and the multi-level semantic ambiguity caused by pooling features over object-level masks, which dilutes fine-grained details. To address these challenges, we present a novel framework that leverages Sparse Voxel Rasterization (SVRaster) as a structured, disjoint geometry representation. By regularizing SVRaster with monocular depth and normal priors, we establish a stable geometric foundation. This enables a deterministic, confidence-aware feature registration process and suppresses the semantic bleeding artifact common in 3DGS. Furthermore, we resolve multi-level ambiguity by exploiting the emerging dense alignment properties of the AM-RADIO foundation model, avoiding the computational overhead of hierarchical training methods. Our approach achieves state-of-the-art performance on Open Vocabulary Point Cloud Understanding, and highly competitive results on 3D and 2D Object Retrieval benchmarks.
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。
