Python cv2.resize() Examples

def resize(video, size, interpolation):
  if interpolation == 'bilinear':
    inter = cv2.INTER_LINEAR
  elif interpolation == 'nearest':
    inter = cv2.INTER_NEAREST
  else:
    raise NotImplementedError

  shape = video.shape[:-3]
  video = video.reshape((-1, *video.shape[-3:]))
  resized_video = np.zeros((video.shape[0], size[1], size[0], video.shape[-1]))
  for i in range(video.shape[0]):
    img = cv2.resize(video[i], size, inter)
    if len(img.shape) == 2:
      img = img[:, :, np.newaxis]
    resized_video[i] = img
  return resized_video.reshape((*shape, size[1], size[0], video.shape[-1])) 

def __call__(self, video):
    """
    Args:
        video (numpy.ndarray): Video to be scaled.
    Returns:
        numpy.ndarray: Rescaled video.
    """
    if isinstance(self.size, int):
      w, h = video.shape[-2], video.shape[-3]
      if (w <= h and w == self.size) or (h <= w and h == self.size):
        return video
      if w < h:
        ow = self.size
        oh = int(self.size*h/w)
        return resize(video, (ow, oh), self.interpolation)
      else:
        oh = self.size
        ow = int(self.size*w/h)
        return resize(video, (ow, oh), self.interpolation)
    else:
      return resize(video, self.size, self.interpolation) 

def __call__(self, video):
    for attempt in range(10):
      area = video.shape[-3]*video.shape[-2]
      target_area = random.uniform(0.08, 1.0)*area
      aspect_ratio = random.uniform(3./4, 4./3)

      w = int(round(math.sqrt(target_area*aspect_ratio)))
      h = int(round(math.sqrt(target_area/aspect_ratio)))

      if random.random() < 0.5:
        w, h = h, w

      if w <= video.shape[-2] and h <= video.shape[-3]:
        x1 = random.randint(0, video.shape[-2]-w)
        y1 = random.randint(0, video.shape[-3]-h)

        video = video[..., y1:y1+h, x1:x1+w, :]

        return resize(video, (self.size, self.size), self.interpolation)

    # Fallback
    scale = Scale(self.size, interpolation=self.interpolation)
    crop = CenterCrop(self.size)
    return crop(scale(video)) 

def step(self, amt=1):
        image = self._capFrame()

        if self.crop:
            image = image[self._cropY + self.yoff:self._ih - self._cropY +
                          self.yoff, self._cropX + self.xoff:self._iw - self._cropX + self.xoff]
        else:
            t, b, l, r = self._pad
            image = cv2.copyMakeBorder(
                image, t, b, l, r, cv2.BORDER_CONSTANT, value=[0, 0, 0])

        resized = cv2.resize(image, (self.width, self.height),
                             interpolation=cv2.INTER_LINEAR)
        if self.mirror:
            resized = cv2.flip(resized, 1)

        for y in range(self.height):
            for x in range(self.width):
                self.layout.set(x, y, tuple(resized[y, x][0:3])) 

def detect(self, img):
        """
        img: rgb 3 channel
        """
        minsize = 20  # minimum size of face
        threshold = [0.6, 0.7, 0.7]  # three steps's threshold
        factor = 0.709  # scale factor

        bounding_boxes, _ = FaceDet.detect_face(
                img, minsize, self.pnet, self.rnet, self.onet, threshold, factor)
        area = (bounding_boxes[:, 2] - bounding_boxes[:, 0]) * (bounding_boxes[:, 3] - bounding_boxes[:, 1])
        face_idx = area.argmax()
        bbox = bounding_boxes[face_idx][:4]  # xy,xy

        margin = 32
        x0 = np.maximum(bbox[0] - margin // 2, 0)
        y0 = np.maximum(bbox[1] - margin // 2, 0)
        x1 = np.minimum(bbox[2] + margin // 2, img.shape[1])
        y1 = np.minimum(bbox[3] + margin // 2, img.shape[0])
        x0, y0, x1, y1 = bbox = [int(k + 0.5) for k in [x0, y0, x1, y1]]
        cropped = img[y0:y1, x0:x1, :]
        scaled = cv2.resize(cropped, (160, 160), interpolation=cv2.INTER_LINEAR)
        return scaled, bbox 

def forward_ocr(self, img_):
        img_ = cv2.resize(img_, (80, 30))
        img_ = img_.transpose(1, 0)
        print(img_.shape)
        img_ = img_.reshape((1, 80, 30))
        print(img_.shape)
        # img_ = img_.reshape((80 * 30))
        img_ = np.multiply(img_, 1 / 255.0)
        self.predictor.forward(data=img_, **self.init_state_dict)
        prob = self.predictor.get_output(0)
        label_list = []
        for p in prob:
            print(np.argsort(p))
            max_index = np.argsort(p)[::-1][0]
            label_list.append(max_index)
        return self.__get_string(label_list) 

def resize(im, short, max_size):
    """
    only resize input image to target size and return scale
    :param im: BGR image input by opencv
    :param short: one dimensional size (the short side)
    :param max_size: one dimensional max size (the long side)
    :return: resized image (NDArray) and scale (float)
    """
    im_shape = im.shape
    im_size_min = np.min(im_shape[0:2])
    im_size_max = np.max(im_shape[0:2])
    im_scale = float(short) / float(im_size_min)
    # prevent bigger axis from being more than max_size:
    if np.round(im_scale * im_size_max) > max_size:
        im_scale = float(max_size) / float(im_size_max)
    im = cv2.resize(im, None, None, fx=im_scale, fy=im_scale, interpolation=cv2.INTER_LINEAR)
    return im, im_scale 

def test_augmenters(self):
        # ColorNormalizeAug
        mean = np.random.rand(3) * 255
        std = np.random.rand(3) + 1
        width = np.random.randint(100, 500)
        height = np.random.randint(100, 500)
        src = np.random.rand(height, width, 3) * 255.
        # We test numpy and mxnet NDArray inputs
        color_norm_aug = mx.image.ColorNormalizeAug(mean=mx.nd.array(mean), std=std)
        out_image = color_norm_aug(mx.nd.array(src))
        assert_almost_equal(out_image.asnumpy(), (src - mean) / std, atol=1e-3)

        # only test if all augmenters will work
        # TODO(Joshua Zhang): verify the augmenter outputs
        im_list = [[0, x] for x in TestImage.IMAGES]
        test_iter = mx.image.ImageIter(2, (3, 224, 224), label_width=1, imglist=im_list,
            resize=640, rand_crop=True, rand_resize=True, rand_mirror=True, mean=True,
            std=np.array([1.1, 1.03, 1.05]), brightness=0.1, contrast=0.1, saturation=0.1,
            hue=0.1, pca_noise=0.1, rand_gray=0.2, inter_method=10, path_root='', shuffle=True)
        for batch in test_iter:
            pass 

def resize_maps(map, map_scales, resize_method):
  scaled_maps = []
  for i, sc in enumerate(map_scales):
    if resize_method == 'antialiasing':
      # Resize using open cv so that we can compute the size.
      # Use PIL resize to use anti aliasing feature.
      map_ = cv2.resize(map*1, None, None, fx=sc, fy=sc, interpolation=cv2.INTER_LINEAR)
      w = map_.shape[1]; h = map_.shape[0]

      map_img = PIL.Image.fromarray((map*255).astype(np.uint8))
      map__img = map_img.resize((w,h), PIL.Image.ANTIALIAS)
      map_ = np.asarray(map__img).astype(np.float32)
      map_ = map_/255.
      map_ = np.minimum(map_, 1.0)
      map_ = np.maximum(map_, 0.0)
    elif resize_method == 'linear_noantialiasing':
      map_ = cv2.resize(map*1, None, None, fx=sc, fy=sc, interpolation=cv2.INTER_LINEAR)
    else:
      logging.error('Unknown resizing method')
    scaled_maps.append(map_)
  return scaled_maps 

def load_image(self, index):
    # loads 1 image from dataset
    img = self.imgs[index]
    if img is None:
        img_path = self.img_files[index]
        img = cv2.imread(img_path)  # BGR
        assert img is not None, 'Image Not Found ' + img_path
        r = self.img_size / max(img.shape)  # size ratio
        if self.augment and r < 1:  # if training (NOT testing), downsize to inference shape
            h, w, _ = img.shape
            img = cv2.resize(img, (int(w * r), int(h * r)), interpolation=cv2.INTER_LINEAR)  # _LINEAR fastest

    # Augment colorspace
    if self.augment:
        augment_hsv(img, hgain=self.hyp['hsv_h'], sgain=self.hyp['hsv_s'], vgain=self.hyp['hsv_v'])

    return img 

def prep_im_for_blob(im, pixel_means, pixel_stds, target_size, max_size):
    """Mean subtract and scale an image for use in a blob."""
    
    im = im.astype(np.float32, copy=False)
    im /= 255.0
    im -= pixel_means
    im /= pixel_stds
    # im = im[:, :, ::-1]
    im_shape = im.shape
    im_size_min = np.min(im_shape[0:2])
    im_size_max = np.max(im_shape[0:2])
    im_scale = float(target_size) / float(im_size_min)
    # Prevent the biggest axis from being more than MAX_SIZE
    # if np.round(im_scale * im_size_max) > max_size:
    #     im_scale = float(max_size) / float(im_size_max)
    # im = imresize(im, im_scale)
    im = cv2.resize(im, None, None, fx=im_scale, fy=im_scale,
                    interpolation=cv2.INTER_LINEAR)

    return im, im_scale 

def grid_batch_images(self, images):
        n, h, w, c = images.shape
        a = int(math.floor(np.sqrt(n)))
        # images = (((images - images.min()) * 255) / (images.max() - images.min())).astype(np.uint8)
        images = images.astype(np.uint8)
        images_in_square = np.reshape(images[:a * a], (a, a, h, w, c))
        new_img = np.zeros((h * a, w * a, c), dtype=np.uint8)
        for col_i, col_images in enumerate(images_in_square):
            for row_i, image in enumerate(col_images):
                new_img[col_i * h: (1 + col_i) * h, row_i * w: (1 + row_i) * w] = image
        resolution = self.cfg.resolution
        if self.cfg.resolution != h:
            scale = resolution / h
            new_img = cv2.resize(new_img, None, fx=scale, fy=scale,
                                 interpolation=cv2.INTER_NEAREST)
        return new_img 

def __getitem__(self, index, to_tensor=True):
    fn = self.image_fns[index]
    img = cv2.cvtColor(cv2.imread(fn, 1), cv2.COLOR_BGR2RGB)

    img, pad_top, pad_left = KuzushijiDataset.pad_to_ratio(img, ratio=1.5)
    h, w = img.shape[:2]
    # print(h / w, pad_left, pad_top)
    assert img.ndim == 3
    scaled_imgs = []
    for scale in self.scales:
      h_scale = int(scale * self.height)
      w_scale = int(scale * self.width)
      simg = cv2.resize(img, (w_scale, h_scale))

      if to_tensor:
        assert simg.ndim == 3, simg.ndim
        simg = simg.transpose((2, 0, 1))
        simg = th.from_numpy(simg.copy())

      scaled_imgs.append(simg)

    return scaled_imgs + [fn] 

def __init__(self, img, use_numpy_fft=True, gauss_kernel=(5, 5)):
        """Constructor

            This method initializes the saliency algorithm.

            :param img: an RGB input image
            :param use_numpy_fft: flag whether to use NumPy's FFT (True) or
                                  OpenCV's FFT (False)
            :param gauss_kernel: Kernel size for Gaussian blur
        """
        self.use_numpy_fft = use_numpy_fft
        self.gauss_kernel = gauss_kernel
        self.frame_orig = img

        # downsample image for processing
        self.small_shape = (64, 64)
        self.frame_small = cv2.resize(img, self.small_shape[1::-1])

        # whether we need to do the math (True) or it has already
        # been done (False)
        self.need_saliency_map = True 

def resized_crop(img, i, j, h, w, size, interpolation='BILINEAR'):
    """Crop the given CV Image and resize it to desired size.

    Args:
        img (CV Image): Image to be cropped.
        i (int): i in (i,j) i.e coordinates of the upper left corner
        j (int): j in (i,j) i.e coordinates of the upper left corner
        h (int): Height of the cropped image.
        w (int): Width of the cropped image.
        size (sequence or int): Desired output size. Same semantics as ``resize``.
        interpolation (int, optional): Desired interpolation. Default is
            ``BILINEAR``.
    Returns:
        CV Image: Cropped image.
    """
    assert _is_numpy_image(img), 'img should be CV Image'
    img = crop(img, i, j, h, w)
    img = resize(img, size, interpolation)
    return img 

def prepare(input):
    # preprocessing the image input
    clean = cv2.fastNlMeansDenoising(input)
    ret, tresh = cv2.threshold(clean, 127, 1, cv2.THRESH_BINARY_INV)
    img = crop(tresh)

    # 40x10 image as a flatten array
    flatten_img = cv2.resize(img, (40, 10), interpolation=cv2.INTER_AREA).flatten()

    # resize to 400x100
    resized = cv2.resize(img, (400, 100), interpolation=cv2.INTER_AREA)
    columns = np.sum(resized, axis=0)  # sum of all columns
    lines = np.sum(resized, axis=1)  # sum of all lines

    h, w = img.shape
    aspect = w / h

    return [*flatten_img, *columns, *lines, aspect] 

def generate_dataset(videos_path, framerate, width, height):
    """Converts videos from specified path to ndarrays of shape [numberOfVideos, -1, width, height, 1]

    Args:
        videos_path: Inside the 'videos/' directory, the name of the subdirectory for videos.
        framerate: The desired framerate of the dataset.
        width: The width we will resize the videos to.
        height: The height we will resize the videos to.

    Returns:
        The dataset with the new size and framerate, and converted to monochromatic.

    """
    dataset = []
    video_index = 0
    for playlist in os.listdir('videos/' + videos_path):
        for video_name in os.listdir('videos/{}/{}'.format(videos_path, playlist)):
            dataset.append([])
            print('Video: {}'.format(video_name))
            video = cv2.VideoCapture('videos/{}/{}/{}'.format(videos_path, playlist, video_name))
            while video.isOpened():
                success, frame = video.read()
                if success:
                    frame = preprocess_image(frame, width, height)
                    dataset[video_index].append(frame)

                    frame_index = video.get(cv2.CAP_PROP_POS_FRAMES)
                    video_framerate = video.get(cv2.CAP_PROP_FPS)
                    video.set(cv2.CAP_PROP_POS_FRAMES, frame_index + video_framerate // framerate)
                    last_frame_index = video.get(cv2.CAP_PROP_FRAME_COUNT)
                    if frame_index >= last_frame_index:
                        # Video is over
                        break

                    
                else:
                    break
            dataset[video_index] = np.reshape(dataset[video_index], (-1, width, height, 1))
            video_index += 1
    return dataset 

def resize_image(img):
    img_size = img.shape
    im_size_min = np.min(img_size[0:2])
    im_size_max = np.max(img_size[0:2])

    im_scale = float(600) / float(im_size_min)
    if np.round(im_scale * im_size_max) > 1200:
        im_scale = float(1200) / float(im_size_max)
    new_h = int(img_size[0] * im_scale)
    new_w = int(img_size[1] * im_scale)

    new_h = new_h if new_h // 16 == 0 else (new_h // 16 + 1) * 16
    new_w = new_w if new_w // 16 == 0 else (new_w // 16 + 1) * 16

    re_im = cv2.resize(img, (new_w, new_h), interpolation=cv2.INTER_LINEAR)
    return re_im, (new_h / img_size[0], new_w / img_size[1]) 

def multiscale_single_test(image, input_scales, predictor):
    '''
    Predict image semantic segmentation labeling using multi-scale inputs.
    Inputs:
    images: numpy array, [height, width, channel], channel = 3.
    input_scales: list of scale factors. e.g., [0.5, 1.0, 1.5].
    predictor: prediction function which takes one scaled image as input and outputs its semantic segmentation labelings.
    Returns:
    Averaged predicted logits of multi-scale inputs
    '''
    image_height_raw = image.shape[0]
    image_width_raw = image.shape[1]
    multiscale_outputs = []
    for input_scale in input_scales:
        image_height_scaled = round(image_height_raw * input_scale)
        image_width_scaled = round(image_width_raw * input_scale)
        image_scaled = cv2.resize(image, (image_width_scaled, image_height_scaled), interpolation=cv2.INTER_LINEAR)
        output = predictor(inputs=[image_scaled], target_height=image_height_raw, target_width=image_width_raw)[0]
        multiscale_outputs.append(output)

    output_mean = np.mean(multiscale_outputs, axis=0)

    return output_mean 

def multiscale_single_validate(image, label, input_scales, validator):

    image_height_raw = image.shape[0]
    image_width_raw = image.shape[1]
    multiscale_outputs = []
    multiscale_losses = []
    for input_scale in input_scales:
        image_height_scaled = round(image_height_raw * input_scale)
        image_width_scaled = round(image_width_raw * input_scale)
        image_scaled = cv2.resize(image, (image_width_scaled, image_height_scaled), interpolation=cv2.INTER_LINEAR)
        output, loss = validator(inputs=[image_scaled], target_height=image_height_raw, target_width=image_width_raw, labels=[label])
        multiscale_outputs.append(output[0])
        multiscale_losses.append(loss)

    output_mean = np.mean(multiscale_outputs, axis=0)
    loss_mean = np.mean(multiscale_losses)

    return output_mean, loss_mean 

def __getitem__(self, index):
        datafiles = self.files[index]
        image = cv2.imread(datafiles["img"], cv2.IMREAD_COLOR)
        image = cv2.resize(image, None, fx=0.5, fy=0.5, interpolation=cv2.INTER_LINEAR)
        size = image.shape
        name = osp.splitext(osp.basename(datafiles["img"]))[0]
        image = np.asarray(image, np.float32)
        image = (image - image.min()) / (image.max() - image.min())
        
        img_h, img_w, _ = image.shape
        pad_h = max(self.crop_h - img_h, 0)
        pad_w = max(self.crop_w - img_w, 0)
        if pad_h > 0 or pad_w > 0:
            image = cv2.copyMakeBorder(image, 0, pad_h, 0, 
                pad_w, cv2.BORDER_CONSTANT, 
                value=(0.0, 0.0, 0.0))
        image = image.transpose((2, 0, 1))
        return image, np.array(size), name 

def prep_im_for_blob(im, pixel_means, target_size, max_size):
  """Mean subtract and scale an image for use in a blob."""
  im = im.astype(np.float32, copy=False)
  im -= pixel_means
  im_shape = im.shape
  im_size_min = np.min(im_shape[0:2])
  im_size_max = np.max(im_shape[0:2])
  im_scale = float(target_size) / float(im_size_min)
  # Prevent the biggest axis from being more than MAX_SIZE
  if np.round(im_scale * im_size_max) > max_size:
    im_scale = float(max_size) / float(im_size_max)
  im = cv2.resize(im, None, None, fx=im_scale, fy=im_scale,
                  interpolation=cv2.INTER_LINEAR)

  return im, im_scale 

def _get_image_blob(im):
  """Converts an image into a network input.
  Arguments:
    im (ndarray): a color image in BGR order
  Returns:
    blob (ndarray): a data blob holding an image pyramid
    im_scale_factors (list): list of image scales (relative to im) used
      in the image pyramid
  """
  im_orig = im.astype(np.float32, copy=True)
  im_orig -= cfg.PIXEL_MEANS

  im_shape = im_orig.shape
  im_size_min = np.min(im_shape[0:2])
  im_size_max = np.max(im_shape[0:2])

  processed_ims = []
  im_scale_factors = []

  for target_size in cfg.TEST.SCALES:
    im_scale = float(target_size) / float(im_size_min)
    # Prevent the biggest axis from being more than MAX_SIZE
    if np.round(im_scale * im_size_max) > cfg.TEST.MAX_SIZE:
      im_scale = float(cfg.TEST.MAX_SIZE) / float(im_size_max)
    im = cv2.resize(im_orig, None, None, fx=im_scale, fy=im_scale,
            interpolation=cv2.INTER_LINEAR)
    im_scale_factors.append(im_scale)
    processed_ims.append(im)

  # Create a blob to hold the input images
  blob = im_list_to_blob(processed_ims)

  return blob, np.array(im_scale_factors) 

def _get_image_blob(im):
  """Converts an image into a network input.
  Arguments:
    im (ndarray): a color image in BGR order
  Returns:
    blob (ndarray): a data blob holding an image pyramid
    im_scale_factors (list): list of image scales (relative to im) used
      in the image pyramid
  """
  im_orig = im.astype(np.float32, copy=True)
  im_orig -= cfg.PIXEL_MEANS

  im_shape = im_orig.shape
  im_size_min = np.min(im_shape[0:2])
  im_size_max = np.max(im_shape[0:2])

  processed_ims = []
  im_scale_factors = []

  for target_size in cfg.TEST.SCALES:
    im_scale = float(target_size) / float(im_size_min)
    # Prevent the biggest axis from being more than MAX_SIZE
    if np.round(im_scale * im_size_max) > cfg.TEST.MAX_SIZE:
      im_scale = float(cfg.TEST.MAX_SIZE) / float(im_size_max)
    im = cv2.resize(im_orig, None, None, fx=im_scale, fy=im_scale,
            interpolation=cv2.INTER_LINEAR)
    im_scale_factors.append(im_scale)
    processed_ims.append(im)

  # Create a blob to hold the input images
  blob = im_list_to_blob(processed_ims)

  return blob, np.array(im_scale_factors) 

def step(self, amt=1):
        ret, frame = self._vid.read()

        image = cv2.cvtColor(frame, cv2.COLOR_RGB2BGRA)

        if self.crop:
            image = image[self._cropY + self.yoff:self._ih - self._cropY +
                          self.yoff, self._cropX + self.xoff:self._iw - self._cropX + self.xoff]
        else:
            t, b, l, r = self._pad
            image = cv2.copyMakeBorder(
                image, t, b, l, r, cv2.BORDER_CONSTANT, value=[0, 0, 0])

        resized = cv2.resize(image, (self.width, self.height),
                             interpolation=cv2.INTER_CUBIC)
        if self.mirror:
            resized = cv2.flip(resized, 1)

        for y in range(self.height):
            for x in range(self.width):
                self.layout.set(x, y, tuple(resized[y, x][0:3]))

        if not isinstance(self.videoSource, int):
            self._frameCount += 1
            if self._frameCount >= self._frameTotal:
                self._vid.set(1, 0)  # CV_CAP_PROP_POS_FRAMES
                self._frameCount = 0
                self.animComplete = True 

def imcv2_affine_trans(im):
	# Scale and translate
	h, w, c = im.shape
	scale = np.random.uniform() / 10. + 1.
	max_offx = (scale-1.) * w
	max_offy = (scale-1.) * h
	offx = int(np.random.uniform() * max_offx)
	offy = int(np.random.uniform() * max_offy)
	
	im = cv2.resize(im, (0,0), fx = scale, fy = scale)
	im = im[offy : (offy + h), offx : (offx + w)]
	flip = np.random.binomial(1, .5)
	if flip: im = cv2.flip(im, 1)
	return im, [w, h, c], [scale, [offx, offy], flip] 

def resize_input(self, im):
	h, w, c = self.meta['inp_size']
	imsz = cv2.resize(im, (w, h))
	imsz = imsz / 255.
	imsz = imsz[:,:,::-1]
	return imsz 

def run(self):
        while True:
            try:
                self.sock.connect(self.ADDR)
                break
            except:
                time.sleep(3)
                continue
        if self.showme:
            cv2.namedWindow('You', cv2.WINDOW_NORMAL)
        print("VEDIO client connected...")
        while self.cap.isOpened():
            ret, frame = self.cap.read()
            if self.showme:
                cv2.imshow('You', frame)
                if cv2.waitKey(1) & 0xFF == 27:
                    self.showme = False
                    cv2.destroyWindow('You')
            sframe = cv2.resize(frame, (0, 0), fx=self.fx, fy=self.fx)
            data = pickle.dumps(sframe)
            zdata = zlib.compress(data, zlib.Z_BEST_COMPRESSION)
            try:
                self.sock.sendall(struct.pack("L", len(zdata)) + zdata)
            except:
                break
            for i in range(self.interval):
                self.cap.read() 

def get_cam(imggrad, conv_out):
    """Compute CAM. Refer section 3 of https://arxiv.org/abs/1610.02391 for details"""
    weights = np.mean(imggrad, axis=(1, 2))
    cam = np.ones(conv_out.shape[1:], dtype=np.float32)
    for i, w in enumerate(weights):
        cam += w * conv_out[i, :, :]
    cam = cv2.resize(cam, (imggrad.shape[1], imggrad.shape[2]))
    cam = np.maximum(cam, 0)
    cam = (cam - np.min(cam)) / (np.max(cam) - np.min(cam)) 
    cam = np.uint8(cam * 255)
    return cam 

def read_image_cv(path):
    return cv2.resize(cv2.cvtColor(cv2.imread(path), cv2.COLOR_BGR2RGB), image_sz)

# synset.txt contains the names of Imagenet categories
# Load the file to memory and create a helper method to query category_index -> category name