外观
论文arXiv 2610.2161913 天前
3D Gaussian Language Map 的免训练目标感知优化
原文链接：Training-Free Target-Aware Refinement of 3D Gaussian Language Map（arXiv 2610.21619 · 发布日期：2026-10-15）
Training-Free Target-Aware Refinement of 3D Gaussian Language Map
提出免训练目标感知优化方法，改进已蒸馏3D Gaussian语言地图，并在三个基准和真机演示中验证。
核心要点
- 已蒸馏语言地图可文本查询，但部署中常漏检用户指定目标，且需保留开放词汇能力。
- 方法由文本提示分割器引导，将目标区域视觉嵌入与目标文本嵌入按闭式系数混合。
- 为避免过度注入破坏外观线索和通用查询性，逐目标求取最小有效系数且无需调参。
- 三个基准上提升开放词汇分割，Replica上mIoU约翻倍，并支持腿式机器人语言目标导航演示。
验证范围与局限：验证包括三个基准、可插入现成场和对象级语言地图，以及一次腿式机器人真实世界演示。
英文摘要 (Abstract)
Distilling vision-language features into 3D Gaussian Splatting yields a language map that can be queried with arbitrary text. In deployment, however, a small set of user-specified targets must be reliably located, which current distilled maps often miss, while open-vocabulary queryability for other words must remain intact. We propose a training-free, target-aware refinement of an already-distilled language map. Guided by a text-promptable segmenter, the visual embedding inside each detected target region is blended with the target text embedding, a convex combination whose coefficient sets the amount of injected prior, while non-target regions are left untouched. Injecting too much overwrites appearance cues and general queryability, so we seek, for each target, the smallest coefficient that still lets it win, and we obtain every coefficient in closed form, leaving no hyperparameters to tune. We further identify a failure mode of injection in language-goal localization. Text embeddings share a common component, and injecting them makes the standard cosine read-out query-agnostic, which a centering step in the read-out removes. Across three benchmarks, the refinement consistently improves open-vocabulary segmentation, roughly doubling the base mIoU on Replica, and localizes language goals more reliably than the base map. It also transfers as a plug-in to off-the-shelf field-based and object-level language maps and supports a real-world demonstration in which a legged robot navigates to natural-language goals. Project page and code: https://lab-of-ai-and-robotics.github.io/target-aware-langmap/
作者：Sibaek Lee, Kyeongsu Kang, Seongbo Ha, Yewon Jeon, Joonyeol Choi, Daegeol Ko, Jaehyuk Hur, Choi Taemin, Minsu Kim, Hyungjoo Ahn
来源
本页事实对应的公开页面，以原文为准。