Pyto includes OpenCV. However, video output does not work the exact same way that on computers. Frames are displayed on the console and no other window is presented. So cv2.destroyAllWindows and cv2.waitKey will throw errors.

Pyto specific API

cv2.autorotate(frame, cam)

Returns an auto rotated frame captured with hardware camera.

By default, frames captured by hardware camera will be rotated correctly only if the device is in portrait mode. By calling this function, you can make sure things like face detection will work on any device orientation.

  • frame – The frame captured by the camera.
  • cam – The camera used to capture the frame. 0 for back and 1 for front.

Face detection example

An example of face detection using OpenCV.

import cv2
import sys

casc_path ="haarcascade_frontalface_default.xml"
face_cascade = cv2.CascadeClassifier(casc_path)

device = 1 # Front camera
    device = int(sys.argv[1]) # 0 for back camera
except IndexError:

cap = cv2.VideoCapture(device)

while cap.isOpened():

    # Capture frame-by-frame
    ret, frame =

    # Check if frame is not empty
    if not ret:

    # Auto rotate camera
    frame = cv2.autorotate(frame, device)

    # Convert from BGR to RGB
    frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)

    faces = face_cascade.detectMultiScale(
      minSize=(30, 30),

    # Draw a rectangle around the faces
    for (x, y, w, h) in faces:
      cv2.rectangle(frame, (x, y), (x+w, y+h), (0, 255, 0), 2)

    # Display the resulting frame
    cv2.imshow('frame', frame)