发表机构
Santa Clara University; University at Albany, SUNY; Iowa State University; Woven by Toyota; United Arab Emirates University; National Yang Ming Chiao Tung University; University of Macau; North Carolina State University; Columbia University; Milestone Systems; Xi’an Jiaotong University; Johns Hopkins University（圣克拉拉大学; 纽约州立大学奥尔巴尼分校; 爱荷华州立大学; 丰田编织公司; 阿拉伯联合酋长国大学; 国立阳明交通大学; 澳门大学; 北卡罗来纳州立大学; 哥伦比亚大学; 里程碑系统公司; 西安交通大学; 约翰斯·霍普金斯大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文总结了与ECCV 2026同期举办的第10届AI City挑战赛的设置、数据集、评估结果等，该赛事规模扩大，设多类赛道，成功系统结合基础模型与多种技术。
AI 中文摘要
第10届AI City挑战赛与2026年欧洲计算机视觉会议（ECCV）同期举办，标志着智能交通、智慧城市与物理人工智能领域的社区基准测试已走过十年历程。自2017年以车辆检测、分类与跟踪项目起步以来，该挑战赛已发展为涵盖多相机感知、多模态推理、合成到真实学习、生成式预测及隐私保护评估的综合基准套件。2026年赛事规模持续扩大，注册参赛团队达325支，较2025年的245支有所增长，参赛国家及地区共26个，较此前的15个进一步增加。本届赛事设6个主要赛道，涵盖多相机3D感知、交通安全字幕与视觉问答（VQA）、交通异常推理、基于文本的人员异常搜索、生成式交通视频预测及跨城市目标检测。第3赛道还新增2个域外排行榜，作为第7、8赛道提交，分别针对鱼眼镜头交通违规理解与行人情境意图视觉问答。本文总结了该挑战赛的设置、数据集、评估协议、排行榜结果及研讨会论文，各赛道的成功系统均结合了基础模型与几何定位、检索或重排序、合成数据设计、域适应及可控推理技术。
英文摘要
The 10th AI City Challenge, held with ECCV 2026, marks a decade of community benchmarking for intelligent transportation, smart cities, and physical AI. Since its 2017 start with vehicle detection, classification, and tracking, the challenge has grown into a broad benchmark suite for multi-camera perception, multimodal reasoning, synthetic-to-real learning, generative forecasting, and privacy-preserving evaluation. The 2026 edition continued this growth with 325 registered teams, up from 245 in 2025, and participation from 26 countries and regions, up from 15. Its six primary tracks cover multi-camera 3D perception, transportation safety captioning and VQA, traffic anomaly reasoning, text-based person anomaly search, generative traffic video forecasting, and cross-city object detection. Track 3 further includes two out-of-domain leaderboards, submitted as Tracks 7 and 8, for fisheye traffic-violation understanding and pedestrian situated-intent VQA. This paper summarizes the challenge setup, datasets, evaluation protocols, leaderboard results, and workshop papers. Across tracks, successful systems combine foundation models with geometric grounding, retrieval or reranking, synthetic-data design, domain adaptation, and controlled inference.
CommentsSummary of the 10th AI City Challenge Workshop in conjunction with ECCV 2026
