isymbo / csofai-session2-project3

Live Streaming and Face Detection with Flask

pip install requirements.txt

Run Server

python app.py

Bring your own models built with Tensorflow

Just put your model under model and protos directory

Use Built-in Webcam of Laptop

Put Zero (O) in cv2.VideoCapture(0)

cv2.VideoCapture(0)
 

Use Ip Camera/CCTV/RTSP Link

cv2.VideoCapture('rtsp://username:password@camera_ip_address:554/user=username_password='password'_channel=channel_number_stream=0.sdp')  

Example RTSP Link

cv2.VideoCapture('rtsp://mamun:123456@101.134.16.117:554/user=mamun_password=123456_channel=0_stream=0.sdp')

Change Channel Number to Change the Camera

cv2.VideoCapture('rtsp://mamun:123456@101.134.16.117:554/user=mamun_password=123456_channel=1_stream=0.sdp')

Display the resulting frame in browser

cv2.imencode('.jpg', frame)[1].tobytes()

Display the resulting frame in window

Instead of return use this in camera.py

cv2.imshow('frame', frame)
if cv2.waitKey(1) & 0xFF == ord('q'):
       break
            

Credit

Learn More about Streaming with flask

• https://blog.miguelgrinberg.com/post/video-streaming-with-flask

Live streaming with face detection

• https://github.com/mrxmamun/live-stream-face-detection

Tensorflow face detector

• https://github.com/yeephycho/tensorflow-face-detection

Models and examples built with TensorFlow

• https://github.com/tensorflow/models