The 10th AI City Challenge
The 10th AI City Challenge was held in conjunction with ECCV 2026. The challenge has grown over the past decade from a focus on vehicle detection and classification to a broader suite of benchmarks for multi-camera perception, multimodal reasoning, and other tasks. This year's edition saw 325 registered teams from 26 countries and regions, up from 245 teams in 2025. The challenge includes six primary tracks and two out-of-domain leaderboards.
The 10th AI City Challenge was held in conjunction with ECCV 2026. The challenge has grown over the past decade from a focus on vehicle detection and classification to a broader suite of benchmarks for multi-camera perception, multimodal reasoning, and other tasks. This year's edition saw 325 registered teams from 26 countries and regions, up from 245 teams in 2025. The challenge includes six primary tracks and two out-of-domain leaderboards.
---
Why it matters: This matters to AI researchers because it provides a benchmark for evaluating the performance of intelligent transportation systems and smart city technologies. Successful systems combine foundation models with geometric grounding, retrieval or reranking, synthetic-data design, domain adaptation, and controlled inference.
Source: https://arxiv.org/abs/2608.17044
This article was originally published at: https://arxiv.org/abs/2608.17044