What datatype is captured with cv2.VideoCapture.read() method in OpenCV?

I am wondering what kind of data type is being captured using the cv2.VideoCapture.read() method. I have been reading the OpenCV documents and I found this explanation:

I have followed a couple of basic tutorials with OpenCV where a webcam can capture data and outputs the frames. In the picture below the frame data is being shown.

And here follows the code that outputs these frames:

import cv2, time, base64

framesCaptured = 0;
video=cv2.VideoCapture(0) <====== Video capture for the webcam

# Save camera frames as movie
fourcc = cv2.VideoWriter_fourcc(*'XVID')
out = cv2.VideoWriter('recording.avi', fourcc, 20.0, (640, 480))

while True:
    framesCaptured += 1

    check, frame = video.read() <====== Grabs and retrieves frames and decode
    # Write video frames
    out.write(frame)

    # Print out statements
    #print(check)
    print(frame) <====== Print out frame data (as shown in the cmd picture below)

    gray=cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)

    cv2.imshow("frame", gray)   

    key=cv2.waitKey(1)

    if key == ord('q'):
        break

#print('Frames captured: ' + str(framesCaptured))
video.release()
out.release()
cv2.destroyAllWindows

So I am wondering what kind of data type is being printed from the 'frame' variable?

In the end I want to use this 'frame' data to re-construct the images to a video in a C# application.