跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2603.09632 · v6

X-GS: An Extensible Framework for Perceiving and Thinking with 3D Gaussian Splatting

首次提交 2026-03-10 21:10
版本更新 2026-11-16 17:39
雷达观测 2026-11-17 17:11
arXiv ↗PDF ↗

摘要

3D Gaussian Splatting (3DGS) has emerged as a powerful technique for novel view synthesis, subsequently extending into numerous spatial AI applications. However, most existing 3DGS methods operate in isolation, focusing on specific domains. In this paper, we introduce X-GS, an extensible framework that integrates previously isolated 3DGS methods into the perception module of a VLM for spatial tasks, with two major components: the Perceiver\textit{Perceiver} and the Thinker\textit{Thinker}. The Perceiver\textit{Perceiver} performs online 3DGS-based SLAM with semantic distillation and outputs semantic Gaussians from unposed video streams. It leverages recent vision foundation models for stronger geometric priors, and we introduce three novel optimizations to improve semantic distillation efficiency. The Thinker\textit{Thinker} interfaces diverse VLMs with these semantic Gaussians, unlocking spatial multimodal capabilities such as 3D visual grounding and scene captioning. Experimental results on diverse benchmarks demonstrate the efficiency and newly unlocked multimodal capabilities of the X-GS framework. Code is available at https://github.com/yueen-ma/X-GS.

作者备注:16 pages, 5 figures. Accepted to Findings of EMNLP 2026

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

本站首次收录为 v6,暂无更早版本可对照。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v6 · 2026-11-16 17:39 · 最新观测版本

X-GS: An Extensible Framework for Perceiving and Thinking with 3D Gaussian Splatting

3D Gaussian Splatting (3DGS) has emerged as a powerful technique for novel view synthesis, subsequently extending into numerous spatial AI applications. However, most existing 3DGS methods operate in isolation, focusing on specific domains. In this paper, we introduce X-GS, an extensible framework that integrates previously isolated 3DGS methods into the perception module of a VLM for spatial tasks, with two major components: the Perceiver\textit{Perceiver} and the Thinker\textit{Thinker}. The Perceiver\textit{Perceiver} performs online 3DGS-based SLAM with semantic distillation and outputs semantic Gaussians from unposed video streams. It leverages recent vision foundation models for stronger geometric priors, and we introduce three novel optimizations to improve semantic distillation efficiency. The Thinker\textit{Thinker} interfaces diverse VLMs with these semantic Gaussians, unlocking spatial multimodal capabilities such as 3D visual grounding and scene captioning. Experimental results on diverse benchmarks demonstrate the efficiency and newly unlocked multimodal capabilities of the X-GS framework. Code is available at https://github.com/yueen-ma/X-GS.

查看此版本解读 →查看此版本原文 ↗