{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/67e87037e4884eab8ab88ee5439de8d4\" frameborder=\"0\" width=\"1920\" height=\"1440\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1440,"width":1920,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1440,"thumbnail_width":1920,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/67e87037e4884eab8ab88ee5439de8d4-d30053152225a22d.gif","duration":843.833333,"title":"Realtime-helmet-Detection-project","description":"In this Loom, I walk you through my real time helmet violation and license plate recognition system. It processes video streams and live phone camera input using a multi stage pipeline, with FastAPI for the async backend and YOLO V8 for detection. I detect four classes, helmet, without helmet, rider, and number plate, then run PaddleOCR plus a regex validator for Indian plate formats, and store violations in files with confidence and timestamps while avoiding duplicates. I also show association logic to map plates to the correct rider and prevent false citations. No specific action was requested from viewers."}