arXiv 2610.26659 · cs.CV · 2026-10-19

SpatialScout: Can VLM Agents Actively Explore and Map 3D Scenes?

Zhaoxi Zhang, Hao Liang, Linxuan Wu, Yiming Xu, Yifu Yuan, Kun Liang, Chenming Tang, Meiyi Qiang 等 10 人

Peking University

在最新榜 #61 本站收录于 2026-10-23

一句话要点

北大发布 SpatialScout 基准：80 个室内场景、3140 个目标物体，要求视觉语言模型智能体自主控制相机探索并估计物体中心坐标，考察选视角、跨视角关联、几何估计与地图修正能力。

Peking University’s SpatialScout benchmark (80 indoor scenes, 3,140 objects) asks VLM agents to steer a camera, explore and estimate object centres in a shared frame, testing viewpoint selection, cross-view association, geometry and map revision.

170HF 点赞

10GitHub stars

0引用（Semantic Scholar）

63.5热度分

阅读 arXiv 原文 PDF Hugging Face 讨论代码仓库项目页

主题与学科

多模态智能体大语言模型

AI × 计算机科学（算法与理论）

同主题：多模态

同主题：智能体

Raven: The Harness of Harnesses for Composable Agentic Intelligence 2026-09-27 · ▲748
False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents 2026-09-30 · ▲642
Omni-IO Skills: Harnessing Your Agent Omni-Native 2026-09-25 · ▲485
BizAgentBench: Unveiling Hidden Challenges in Deploying Long-Horizon Agents for Real-World Business 2026-10-14 · ▲388
Any Time, Any Space: A World-State-Centric Framework for Agentic Video Generation 2026-10-14 · ▲364

同机构：Peking University

Open-o3-Video-v2: Seek, Ground, and Reason with Verifiable Spatio-Temporal Evidence 2026-10-13
Duplex-MPE: Benchmarking Multi-Party Interaction in Full-Duplex Dialogue 2026-09-25
Post-Training Leaves Behavioral Shadows on Unrelated Decisions 2026-09-24
Harness-Zero: Harness Distillation via Agent-as-Harness 2026-09-21
OmniEdu: Open Foundation Models for Learning and Teaching 2026-09-19

数据说明

元数据来自 arXiv API（已核实）；点赞来自 Hugging Face 每日论文；引用来自 Semantic Scholar。最近刷新 2026-10-26。

要点由本站根据摘要自写，不转载摘要原文，请阅读原文。