Usage:-

• Clone this repository.
• Open cmd in working directory.
• Execute following command:
  • pip install -r requirements.txt
• FACE_detect.py is the file which contains main function of the Face detection and counter.
• app.py is the main python file for Streamlit Web Application.
• To run app, write following command in CMD. or use any IDE. (This will run the app on localhost)
  • streamlit run app.py
• Note: The webapp feature for deployed app may not work properly as streamlit-webrtc is still under creation. It is difficult to create live stream computer vision apps on streamlit as this means to refresh the streamlit large number of times in a short period of time.

################ Information about other work done in this repository ##################

This Repository contains programs to detect objects using OpenCV.

Haarcascades XML files are used to create claasifiers. All xml files are stored in /Haarcascades directory. Dataset link : https://www.kaggle.com/lalitharajesh/haarcascades

Some image and video examples are saved in /image_examples directory for testing purposes.

1. app.py

   • This file contains face detection and counting code.
   • It is also deployed(Very basic version) using streamlit.
   • User can upload any image and detect faces. Results can also be downloaded in image format.

2. car_detection.py

   • Detects cars in a video.
   • Rectangles are created around the video.

3. face_and_eye_detection.py

   • Detects face as well as eyes from a picture.
   • It can be also modified for live videos from webcam or any other camera source.

4. face_and_eye_detection_using_webcam.py

   • Detects live faces and eyes from a live webcam.

5. full_body_detection.py

   • Detects full body from a video source file.
   • It can be used as pedestrian detection tool or modified to some social distancing verifier.

6. something_very_cool.py

   • This program lets user to draw a rectangle on the first frame captured by webcam.
   • Then this frame remains in sketch format.
   • It looks really cool...just try it out.

NOTE:- To close/quit cv2 frames press- Enter or 'q'.