- 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 ##################
Haarcascades XML files are used to create claasifiers. All xml files are stored in /Haarcascades directory. Dataset link : https://www.kaggle.com/lalitharajesh/haarcascades
-
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.
-
car_detection.py
- Detects cars in a video.
- Rectangles are created around the video.
-
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.
-
face_and_eye_detection_using_webcam.py
- Detects live faces and eyes from a live webcam.
-
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.
-
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.