A Review Paper on Social Distance Detection Using Deep Learning


  • Shinde Raj Rahul
  • Deshmukh Mangesh Prakash
  • Shikhare Kedar Shashikant
  • Kale Abhishek
  • Kiran Jagtap


CNN, Logistic Regression, Feature Extraction, Early blight, Late blight.


In order to lessen the effects of this coronavirus, the research provides a mechanism for social distance identification using deep learning. Pandemic. By analyzing a video feed, the detecting tool was created to warn individuals to keep a safe distance from one another. The open-source object detection pre- trained model based on the YOLOv3 method was used to detect pedestrians using the video frame from the camera as input. Later, a top-down perspective of the video frame was added to estimate distances from the 2D plane. It is possible to estimate the distance between individuals, and any non-compliant pair of individuals in the display will be denoted by a red frame and red line. A recorded video of people crossing the street was used to validate the proposed method. The outcome demonstrates that the suggested method is capable of identifying the social distances between various characters in the video. The developed method can be improved and used as a real-time detection tool. Keywords – pedestrian detection, social distance, deep learning, and convolutional neural network.


Download data is not yet available.




How to Cite

Shinde Raj Rahul, Deshmukh Mangesh Prakash, Shikhare Kedar Shashikant, Kale Abhishek, & Kiran Jagtap. (2023). A Review Paper on Social Distance Detection Using Deep Learning. International Journal of Progressive Research in Science and Engineering, 4(5), 247–250. Retrieved from https://journal.ijprse.com/index.php/ijprse/article/view/870