ReLaGS：关系语言高斯点云
ReLaGS: Relational Language Gaussian Splatting
AI总结：
ReLaGS通过构建层次化的语言蒸馏高斯场景和3D语义场景图，实现无需场景特定训练的统一3D感知与推理，结合多视角语言对齐策略和图神经网络进行关系推理。
摘要：
实现跨分割、检索和关系理解等任务的统一3D感知与推理仍然具有挑战性，因为现有方法要么以对象为中心，要么依赖于昂贵的训练进行跨对象推理。我们提出了一种新的框架，构建了层次化的语言蒸馏高斯场景及其3D语义场景图，而无需场景特定的训练。高斯修剪机制优化场景几何，而稳健的多视角语言对齐策略将噪声的2D特征聚合为准确的3D对象嵌入。在此层次之上，我们构建了一个开放词汇的3D场景图，利用视觉语言推导注释和基于图神经网络的关系推理。我们的方法通过联合建模层次语义和跨/内对象关系，实现了高效且可扩展的开放词汇3D推理，已在分割、场景图生成和关系引导检索等任务中得到验证。项目页面：https://dfki-av.github.io/ReLaGS/
英文摘要：
Achieving unified 3D perception and reasoning across tasks such as segmentation, retrieval, and relation understanding remains challenging, as existing methods are either object-centric or rely on costly training for inter-object reasoning. We present a novel framework that constructs a hierarchical language-distilled Gaussian scene and its 3D semantic scene graph without scene-specific training. A Gaussian pruning mechanism refines scene geometry, while a robust multi-view language alignment strategy aggregates noisy 2D features into accurate 3D object embeddings. On top of this hierarchy, we build an open-vocabulary 3D scene graph with Vision Language derived annotations and Graph Neural Network-based relational reasoning. Our approach enables efficient and scalable open-vocabulary 3D reasoning by jointly modeling hierarchical semantics and inter/intra-object relationships, validated across tasks including open-vocabulary segmentation, scene graph generation, and relation-guided retrieval. Project page: https://dfki-av.github.io/ReLaGS/
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。
