{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/e37b72f995e64aca98a3381269a3a3aa\" frameborder=\"0\" width=\"1670\" height=\"1252\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1252,"width":1670,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1252,"thumbnail_width":1670,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/e37b72f995e64aca98a3381269a3a3aa-e11fde1fbfecd5ae.gif","duration":199.253,"title":"Cyclist Safety System for Near Incidents","description":"This Loom discusses a near-miss and collision detection approach for rider safety using data inputs rather than video cameras. The team set out to identify issues from limited cycle data, marking incidents and compiling them into coaching materials that help riders improve how they share the same path. They use technologies including Cosmo for measuring distances, read compile windows, and generate reports from video sets using an input file path. The author notes they could improve collision detection where models did not distinguish whether the car moved into the cyclist path or vice versa, and they propose building a dash camera capability using drivers dash cams in the future."}