Add possibility to process output of VideoCapture::read

[DamianMel]

Hi!
I very like your project, great job!

I'm thinking about one feature which would be very useful:
OpenCV allows to grab frame after frame from realitme stream (https://docs.opencv.org/2.4/modules/highgui/doc/reading_and_writing_images_and_video.html#videocapture-read)

For now workaround would be to read frame, save it to file and read img file via PassportEye.

Could it be possible for you to add such feature?

Have a nice day.

[DamianMel]

I recently found workaround for this issue:

is_success, buffer = cv2.imencode(".jpg", img)
io_buf = io.BytesIO(buffer)
mrz = read_mrz(io_buf)

Maybe it helps someone

[konstantint]

Hey, sorry for a delayed reaction. Good that you found the workaround.
In general it is true that forcing a file (or a file stream) as input is an unnecessary limitation, and supporting raw pixel arrays as input might be nice as well. I'll try to add this feature when I get the time to work on the code again.