Social Distancing Checker using DL using pre-trained YOLOv4 model !

Developed an Artificial Intelligence powered tool using YOLO object detection model, DeepSORT tracking algorithm, triggering alerts (red bounding box) who violates COVID 50 pixels distance calculated via individual bounding box centroid.

Detect the social distance violations!

"Social distancing is the only way to counter Covid-19"

Requirements (Technologies used)

• Python
• DeepSORT (For object tracking)
• Object Detection models (Used YOLO Pretrained model for object Detection-(used filter for people indentification only))

RESULT OF MODEL ON TESTING

1. Video1 Output

https://github.com/akshupande/Social-Distancing-Detector/blob/2fc5c0c91f7a0c3038d32e9917570ffaa4fccba4/Output/output.avi

2. Video2 Output

https://github.com/akshupande/Social-Distancing-Detector/blob/2fc5c0c91f7a0c3038d32e9917570ffaa4fccba4/Output/output2.mp4

Effective in surveillance for catching social distancing Violators.

⏳ Future upgrades:

Features	Status
Camera calibration for intrinsic parameter (distance)
Add facemask detection
upgrade to YOLO v7

Kindly check out below links:

👀 YOLO Pre-Trained Model

📊 Dataset
MegaPixels: Origins, Ethics, and Privacy Implications of Publicly Available Face Recognition Image Datasets