ECCV Paper Introduces 360CityArena: A Realistic Urban Navigation Benchmark
CSProfKGD · x · 2026-08-13
Researchers have introduced 360CityArena for ECCV 2026, a realistic urban navigation benchmark built from 360° videos of Akihabara, Tokyo. It aims to evaluate whether embodied AI agents can understand, navigate, and reason about real-world urban environments.
The benchmark includes 602 real-world videos covering 85 streets, featuring 175 navigation and spatial reasoning tasks. The results highlight a massive gap between current AI and human capabilities: while humans achieved a 77.3% success rate, the best LMM-based agent scored only 17.1%. The entire environment and all tasks are fully open-sourced.
Related event: HAL-UTokyo Releases 360CityArena City-Level Navigation Benchmark(2 posts)→
More from Embodied
- DynaRobotics CEO Lindon Gao on running deployment and research feedback loops together — micoolcho · 2026-10-16
- AutoAgent0: VLM-powered agentic runtime lifts long-tail driving success by 28.3pp — yining_hong · 2026-10-16
- CarryOn: dual-loop harness adds personalization and safe recovery to frozen VLA robots — Zixuan Wang · 2026-10-16
- FoveaVLA replaces wrist cameras with learned foveation, hitting 70.8% success — Haoyu_Xiong_ · 2026-10-16
- Musk teases two-way Grok x Optimus integration: Grok in the bot, bot in Grok — beffjezos · 2026-10-16
- SimVTA uses simulator-rendered video to post-train driving world action models with RL — OpenDriveLab · 2026-10-16