AI models rework traffic videos for safer roads

New York City’s thousands of traffic cameras record endless hours of footage every day, but analyzing those videos to identify safety issues and implement improvements typically requires resources that most transportation companies don’t have.

Now researchers at NYU’s Tandon School of Engineering have developed an artificial intelligence system that can automatically detect collisions and near misses in existing traffic videos by combining linguistic reasoning and visual intelligence, potentially changing the way cities improve traffic safety without major new investments.

The study, published in the journal Accident Analysis and Prevention, received the City of New York’s Vision Zero Research Award, an annual recognition of work that aligns with the city’s traffic safety priorities and provides actionable insights. Professor Kaan Ozbay, the paper’s senior author, presented the study on November 19 at the eighth annual Research on the Road symposium.

The work is an example of interdisciplinary…