Python • Running of program gets stuck on tensorflow code

Code:

import cv2from time import timefrom picamera2 import Picamera2from tflite_support.task import corefrom tflite_support.task import processorfrom tflite_support.task import visionimport utilsmodel = "efficientdet_lite0.tflite"    numThreads = 4    dispW = int(1280 / 2)    dispH = int(720 / 2)    picam2 = Picamera2()    picam2.preview_configuration.main.size=(dispW, dispH)    picam2.preview_configuration.main.format='RGB888'    picam2.preview_configuration.align()    picam2.configure("preview")    picam2.start()    pos = (20,60)    font = cv2.FONT_HERSHEY_SIMPLEX    height = 1.5    weight = 3    myColor = (255, 0, 0)    fps = 0    baseOptions = core.BaseOptions(file_name=model, use_coral=False, num_threads=numThreads)    detectionOptions = processor.DetectionOptions(max_results=3, score_threshold=0.5)    options = vision.ObjectDetectorOptions(base_options=baseOptions, detection_options=detectionOptions)    detector = vision.ObjectDetector.create_from_options(options)    while True:        tStart = time()        im = picam2.capture_array()        im = cv2.flip(im, -1)        imRGB = cv2.cvtColor(im, cv2.COLOR_BGR2RGB) # convert from BGR to RGB image        imTensor = vision.TensorImage.create_from_array(imRGB) # create a tensor image        myDetections = detector.detect(imTensor) # get the objects that are detected by tensorflow        image = utils.visualize(im, myDetections) # create a decorated image with detected objects        cv2.putText(im, str(int(fps)) + " FPS", pos, font, height, myColor, weight)        cv2.imshow('Camera', im)        if cv2.waitKey(1)==ord('q'):            break        tEnd = time()        loopTime = tEnd - tStart        fps = 0.9 * fps + 0.1 * 1/loopTime    cv2.destroyAllWindows()

Code:

myDetections = self._detector.detect(imTensor)

Code:

import cv2import osos.environ["LIBCAMERA_LOG_LEVELS"] = "3" #disable info and warning loggingfrom picamera2 import Picamera2from time import timefrom tflite_support.task import corefrom tflite_support.task import processorfrom tflite_support.task import visionimport utilsclass Camera:    def __init__(self, resolution, rotation=True):        self._dispW, self._dispH = resolution        self._centerX = int(self._dispW / 2)        self._centerY = int(self._dispH / 2)        self._rotation = rotation        self._picam2 = None        # text on video properties        self._colour = (0, 255, 0)        self._textPositions = self._set_text_positions()        self._font = cv2.FONT_HERSHEY_SIMPLEX        self._scale = 1        self._thickness = 1        self._angleText = None        self._speedText = None        self._turnText = None        self._zoomValue = 1.0        self._hudActive = True        self._carEnabled = False        self._servoEnabled = False        self._fps = 0        self._weightPrevFps = 0.9        self._weightNewFps = 0.1        self._fpsPos = (10, 30)        self._arrayDict = None        self._number_to_turnValue = {            0: "-",            1: "Left",            2: "Right"        }        #tensorflow variables        model = "/home/christian/Python/RoboCar_2.0/src/efficientdet_lite0.tflite" #needs to be full path #TODO: move to config file        numThreads = 4        baseOptions = core.BaseOptions(file_name=model, use_coral=False, num_threads=numThreads)        detectionOptions = processor.DetectionOptions(max_results=3, score_threshold=0.5)        options = vision.ObjectDetectorOptions(base_options=baseOptions, detection_options=detectionOptions)        self._detector = vision.ObjectDetector.create_from_options(options)    def setup(self):        self._picam2 = Picamera2()        self._picam2.preview_configuration.main.size = (self._dispW, self._dispH)        self._picam2.preview_configuration.main.format = 'RGB888'        self._picam2.preview_configuration.align()        self._picam2.configure("preview")        self._picam2.start()    def show_camera_feed(self, shared_array):        tStart = time() # start timer for calculating fps        # get raw image        im = self._picam2.capture_array()        if self._rotation:            im = cv2.flip(im, -1)        #TODO: wrap this in a method        imRGB = cv2.cvtColor(im, cv2.COLOR_BGR2RGB)  # convert from BGR to RGB image        imTensor = vision.TensorImage.create_from_array(imRGB)  # create a tensor image        # This is the line that the code gets stuck on        myDetections = self._detector.detect(imTensor)  # get the objects that are detected by tensorflow        #image = utils.visualize(im, myDetections)  # create a decorated image with detected objects        # read control values from external classes        self._read_control_values_for_video_feed(shared_array)        # resize image when zooming        if self._zoomValue != 1.0:            im = self._get_zoomed_image(im)        # add control values to cam feed        if self._hudActive:            self._add_text_to_cam_feed(im)        cv2.imshow("Camera", im)        cv2.waitKey(1)        # calculate fps        self._calculate_fps(tStart)    def cleanup(self):        cv2.destroyAllWindows()        self._picam2.close()    def set_car_enabled(self):        self._carEnabled = True    def set_servo_enabled(self):        self._servoEnabled = True    def add_array_dict(self, arrayDict):        self._arrayDict = arrayDict    def _set_text_positions(self):        spacingVertical = 30        horizontalCoord = 10        verticalCoord = self._dispH - 15        positions = []        for i in range(4):            position = (horizontalCoord, verticalCoord - i * spacingVertical)            positions.append(position)        return positions    def _calculate_fps(self, startTime):        endTime = time()        loopTime = endTime - startTime        self._fps = self._weightPrevFps * self._fps + self._weightNewFps * (1 / loopTime)    def _get_zoomed_image(self, image):        halfZoomDisplayWidth = int(self._dispW / (2 * self._zoomValue))        halfZoomDisplayHeight = int(self._dispH / (2 * self._zoomValue))        regionOfInterest = image[self._centerY - halfZoomDisplayHeight:self._centerY + halfZoomDisplayHeight,                           self._centerX - halfZoomDisplayWidth:self._centerX + halfZoomDisplayWidth]        im = cv2.resize(regionOfInterest, (self._dispW, self._dispH), cv2.INTER_LINEAR)        return im    def _add_text_to_cam_feed(self, image):        # add control values to camera feed        counter = 0        cv2.putText(image, "Zoom: " + str(self._zoomValue) + "x", self._get_origin(counter), self._font, self._scale,                    self._colour,                    self._thickness)        counter += 1        if self._angleText:            cv2.putText(image, self._angleText, self._get_origin(counter), self._font, self._scale, self._colour,                        self._thickness)            counter += 1        if self._speedText:            cv2.putText(image, self._speedText, self._get_origin(counter), self._font, self._scale, self._colour,                        self._thickness)            counter += 1        if self._turnText:            cv2.putText(image, self._turnText, self._get_origin(counter), self._font, self._scale, self._colour,                        self._thickness)            counter += 1        # display fps        cv2.putText(image, self._get_fps(), self._fpsPos, self._font, self._scale, self._colour,                    self._thickness)    def _get_fps(self):        return str(int(self._fps)) + " FPS"    def _read_control_values_for_video_feed(self, shared_array):        if self._carEnabled:            self._speedText = "Speed: " + str(int(shared_array[self._arrayDict["speed"]])) + "%"            self._turnText = "Turn: " + self._get_turn_value(shared_array[self._arrayDict["turn"]])        if self._servoEnabled:            self._angleText = "Angle: " + str(int(shared_array[self._arrayDict["servo"]]))        self._hudActive = shared_array[self._arrayDict["HUD"]]        self._zoomValue = shared_array[self._arrayDict["Zoom"]]    def _get_turn_value(self, number):        return self._number_to_turnValue[number]    def _get_origin(self, count):        return self._textPositions[count]