Executive Summary
Social Distancing has been proven measure in preventing the spread of infectious diseases such as COVID-19 but it has been the most difficult to track and enforce. IP Cameras powered by AI vision could be one of the accurate & robust solutions to detect if people are at a specific distance and adhere to social distancing or not.
eInfochips has developed social distancing detection powered by NVIDIA® Jetson™ AGX Xavier. Videos from the IP cameras are captured, decoded through RTP client, and pre-processed. Once it’s pre-processed, the pre-trained Deep Learning model detects people information from video frames & then, an algorithm developed by our engineers, extract insights related to people density/social distancing in real-time. These insights are stored locally on the edge and transferred to the server or cloud, which will be accessible to concerned authorities or security managers to review social distancing & take necessary actions in case of any violations.
Project Highlights
- Algorithm Development & Porting for Social Distancing Detection
- Powered by NVIDIA® Jetson AGX Xavier™, Metropolis, Transfer Learning Toolkit (PeopleNet)
- Edge to Cloud Integration (Azure Cloud)
- Rule Based Alerts