aiInternal Project
Social Distance Detection
Computer vision application that measures social distancing violations with 93% accuracy using real-time video analysis.
Developed at ProManage
Lead Data Scientist
2020
Internal project - no public access
Screenshots
About This Project
Created a social distancing measurement application that detects distance violations between people in real-time using computer vision and deep learning. The system processes video feeds to identify individuals, calculate inter-person distances using calibrated camera perspectives, and flag violations to promote public safety compliance.
Key Features
- Real-time person detection using deep learning models
- Distance measurement with camera calibration
- Violation flagging with visual alerts and overlays
- Multi-camera support for large areas
- Statistical reporting on violation patterns
- Configurable distance thresholds and sensitivity
Challenges & Solutions
- Accurately estimating real-world distances from 2D camera feeds
- Handling occlusion and overlapping people in crowded scenes
- Maintaining real-time performance on standard hardware
- Adapting to varying camera angles and perspectives
Results & Impact
- 93% accuracy in social distance violation detection
- Real-time processing capability on standard hardware
- Deployed to promote public safety compliance
Technologies Used
PythonTensorFlowOpenCVDeep LearningNumPyDocker
Project Details
- Category
- ai
- My Role
- Lead Data Scientist
- Duration
- 4 months
- Year
- 2020
- Company
- ProManage
- Status
- Internal / Private
Internal Project
This project was developed for internal use at ProManage. Source code and live demo are not publicly available due to confidentiality.