socialdistance_dnn

Problem Statement: Social distancing analysis and finding no of people in Risk.

Objectives

• Explore the option for video analysis.
• Find the no of people in the video and identify nearby people not maintaining social distance.

Prerequisites

Need to install following:

Python3
opencv
opencv-contrib

Download yolov3 weights from [yolov3_weights](https://pjreddie.com/media/files/yolov3.weights).

Running the project

python[3] main.py
python[3] f2vid.py

Output

* A list of frames in frames and offenders folders respectively.
* Video formed from list of frames.

main folder output: Total no.of frames: 194   Total no.of Offending people: 60

Acknowledgments

• YOLOv3
• Running deep learning models in OpenCV
• yolo-object-detection-opencv-python
• OpenCV Documentation
• Non-Maximum Suppression for Object Detection in Python
• METHODS FOR MEASURING DISTANCE IN IMAGES