@mikael Could you share some picture of sudoku, where recognition fails?

@JonB 350 passes and still works. I'll play with parameters to see how it will go.

Thanks.

This one works: link

from ctypes import *
from objc_util import *

class CMTime(Structure):
  _fields_ = [('value', c_int64), ('timescale', c_int32), ('flags', c_uint32), ('epoch', c_int64)]

AVCaptureDevice = ObjCClass('AVCaptureDevice')

_device = AVCaptureDevice.defaultDeviceWithDeviceType_mediaType_position_(
  'AVCaptureDeviceTypeBuiltInWideAngleCamera', 'vide', 1)

dur = _device.activeVideoMinFrameDuration(restype=CMTime, argtypes=[])
dur.value *= 3
_device.lockForConfiguration_(None)
_device.setActiveVideoMinFrameDuration_(dur, restype=None, argtypes=[CMTime])
_device.unlockForConfiguration()

It seems that setMinFrameDuration is in camera.captureOutput object ('AVCaptureVideoDataOutput').

But there is also setActiveVideoMinFrameDuration in camera.captureDevice ('AVCaptureDevice').

I couldn't make them working too both.

@JonB I have tested another good example here

In this script is implemented

self.captureOutput.setMinFrameDuration_(CMTimeMake(1, 2), argtypes=[CMTime], restype=None)

I have plaeyd with CMTimeMake(.,.).

Unfotunately this script also hangs.

@JonB For data processing.

Other way of obtaining video camera frames continuously is welcome.

@JonB Tried, but same behaviour.

When phone wakes up after auto lock, script makes another 80-90 passes and hangs again.

@mikael Correct duration should be 80-90 passes of face detector.