New AI Model Helps Cities Analyse Traffic Risks at Scale
Researchers at NYU Tandon’s C2SMART developed SeeUnsafe, a multimodal AI system that analyzes traffic-camera footage to detect collisions and near-misses, helping cities draw safety insights from existing video networks. The system won New York City's Vision Zero Research Award. Professor Kaan Ozbay, the study's senior author, explained: "Cities have rapidly expanded their traffic-camera networks, but their ability to learn from all that video hasn't kept up. Manual video analysis is extremely time-consuming and expensive."
