{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/c606ed02018e4a7ea7988e8ae457cc92\" frameborder=\"0\" width=\"1728\" height=\"1296\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1296,"width":1728,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1296,"thumbnail_width":1728,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/c606ed02018e4a7ea7988e8ae457cc92-84de5e364697f42e.gif","duration":234.837,"title":"Dari Sentinel, Real-Time Crash Detection and Rerouting","description":"This Loom explains how Dari Sentinel improves highway incident detection beyond basic map alerts. It uses a road camera and its system to confirm real crashes, identify vehicle details like a multi vehicle crash with an overturned truck, estimate how long it will take to clear, and reroute travelers, with an example showing delays narrowing from 60 to 30 minutes and then to 10 minutes. The creator tested Dari Sentinel on 20 real clips and reported 19 out of 19 incident detections with zero false alarms on normal traffic, using cues like evidence photos and confidence levels with an estimated 47 to 87 minute clearance range. The pipeline involves dashcam or traffic cam detection, a quick incident reality check, SAM 3.1 and Muse Park verification, and Muse Wonder time estimation."}