Case study · 2020
Social Distancing Monitoring System
Real-time, multi-camera system that flags social-distancing violations from existing CCTV feeds.
40+
CCTV cameras monitored
Real-time
Violation alerts
1
Patent filed (India)
Context
Blackstraw.ai, 2020. An end-to-end video-analytics system for a facility's existing CCTV network.
Problem
The facility needed social-distancing violations flagged in real time across its 40+ CCTV cameras, using the cameras already installed on site.
Approach
The system detects people in each feed, estimates the real-world distance between them and flags violations of social-distancing rules in real time.
It is built on NVIDIA DeepStream with TensorRT-optimised detectors and GStreamer/FFmpeg video pipelines, packaged with Docker for deployment on site.
Results
- Deployed at a facility with 40+ CCTV cameras
- Patent filed in India (202021034381)
Stack
- NVIDIA DeepStream
- TensorRT
- GStreamer
- FFmpeg
- Docker
Have a similar problem?
Tell me what you're working on and I'll tell you honestly whether I can help.