The 10th AI City Challenge, held in conjunction with ECCV 2026, has concluded its tenth year of benchmarking intelligent transportation and smart city technologies. This year saw increased participation with 325 registered teams from 26 countries, up from 245 teams in 2025. The challenge featured six main tracks focusing on areas such as multi-camera 3D perception, transportation safety, traffic anomaly reasoning, and generative traffic video forecasting, with additional tracks for specialized tasks like fisheye camera analysis and pedestrian intent. AI
IMPACT This challenge advances benchmarks for AI in intelligent transportation and smart cities, pushing research in areas like multi-camera perception and generative forecasting.
RANK_REASON The item describes a research challenge and its outcomes, including datasets, evaluation protocols, and leaderboard results. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
