Hi, i'm very thankful for the developer of this project i was looking for working video style transfer ai for many months, i was literally dying finding it Thank you a lot ♥️

I have question about resolution, i have resources and time to make 1080p But when i change frame height, it makes the style_ref images smaller on the result Is there a way to keep the size of the pattern same as height 400, but make 1080 resolution, thank you 😃

I receive this error when PRESERVE_COLORS = True. Tried mucking around to change array types but it looks like that should already be taken care of with .astype(np.uint8). Otherwise amazing program, works like a charm on my M1 Macbook Pro. Thanks in advance.

(tf_m1) anthony@Anthonys-MacBook-Pro ~ % python /Users/anthony/Documents/style-transfer-video-processor-main/style_frames.py Getting input frames Getting style info Getting output frames Output frame: 0% Traceback (most recent call last): File "/Users/anthony/Documents/style-transfer-video-processor-main/style_frames.py", line 211, in StyleFrame().run() File "/Users/anthony/Documents/style-transfer-video-processor-main/style_frames.py", line 206, in run self.get_output_frames() File "/Users/anthony/Documents/style-transfer-video-processor-main/style_frames.py", line 167, in get_output_frames temp_ghost_frame = cv2.cvtColor(ghost_frame, cv2.COLOR_RGB2BGR) * self.MAX_CHANNEL_INTENSITY cv2.error: OpenCV(4.5.1) ../modules/imgproc/src/color.simd_helpers.hpp:94: error: (-2:Unspecified error) in function 'cv::impl::(anonymous namespace)::CvtHelper<cv::impl::(anonymous namespace)::Set<3, 4, -1>, cv::impl::(anonymous namespace)::Set<3, 4, -1>, cv::impl::(anonymous namespace)::Set<0, 2, 5>, cv::impl::(anonymous namespace)::NONE>::CvtHelper(cv::InputArray, cv::OutputArray, int) [VScn = cv::impl::(anonymous namespace)::Set<3, 4, -1>, VDcn = cv::impl::(anonymous namespace)::Set<3, 4, -1>, VDepth = cv::impl::(anonymous namespace)::Set<0, 2, 5>, sizePolicy = cv::impl::(anonymous namespace)::NONE]'

Unsupported depth of input image: 'VDepth::contains(depth)' where 'depth' is 6 (CV_64F)

Traceback (most recent call last): File "style_frames.py", line 112, in <module> sf.get_style_info() File "style_frames.py", line 57, in get_style_info self.t_const = np.ceil(frame_length / (ref_count - 1)) ZeroDivisionError: division by zero

I'm a newbie. Can you please make readme a lil bit easy to read for a newbie. 😅 Great repo btw. I'm using only one style image. Here is the colab you can replicate this or can add this to the repo as well.

My config.py looks like this:

    INPUT_VIDEO_NAME = 'input.mp4'
    INPUT_VIDEO_PATH = f'{ROOT_PATH}/{INPUT_VIDEO_NAME}'
    INPUT_FRAME_DIRECTORY = f'{ROOT_PATH}/input_frames'
    INPUT_FRAME_FILE = '{:0>4d}_frame.png'
    INPUT_FRAME_PATH = f'{INPUT_FRAME_DIRECTORY}/{INPUT_FRAME_FILE}'

    STYLE_REF_DIRECTORY = f'{ROOT_PATH}/style_ref'
    # defines the reference style image transition sequence. Values correspond to indices in STYLE_REF_DIRECTORY
    STYLE_SEQUENCE = [1]

    OUTPUT_FPS = 30
    OUTPUT_VIDEO_NAME = 'output_video.mp4'
    OUTPUT_VIDEO_PATH = f'{ROOT_PATH}/{OUTPUT_VIDEO_NAME}'
    OUTPUT_FRAME_DIRECTORY = f'{ROOT_PATH}/output_frames'
    OUTPUT_FRAME_FILE = '{:0>4d}_frame.png'
    OUTPUT_FRAME_PATH = f'{OUTPUT_FRAME_DIRECTORY}/{OUTPUT_FRAME_FILE}'

    GHOST_FRAME_TRANSPARENCY = 0.1`

Also this error too:

`Traceback (most recent call last):
  File "style_frames.py", line 114, in <module>
    sf.get_output_frames()
  File "style_frames.py", line 84, in get_output_frames
    blended_img = prev_style + next_style
ValueError: operands could not be broadcast together with shapes (1138,1707,3) (1138,1713,3) `

error:

`TF Version: 2.4.1 TF Hub version: 0.11.0 Eager mode enabled: True GPU available: [PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')] Getting input frames Getting style info WARNING: Resizing style images which may cause distortion. To avoid this, please provide style images with the same dimensions Getting output frames Output frame: 0% Traceback (most recent call last): File "style_frames.py", line 216, in StyleFrame().run() File "style_frames.py", line 211, in run self.get_output_frames() File "style_frames.py", line 154, in get_output_frames stylized_img = self.hub_module(expanded_content_img, expanded_blended_img).pop() File "/home/chop/Scripts/style-transfer-video-processor/.venv/lib/python3.8/site-packages/tensorflow/python/saved_model/load.py", line 668, in _call_attribute return instance.call(*args, **kwargs) File "/home/chop/Scripts/style-transfer-video-processor/.venv/lib/python3.8/site-packages/tensorflow/python/eager/function.py", line 1669, in call return self._call_impl(args, kwargs) File "/home/chop/Scripts/style-transfer-video-processor/.venv/lib/python3.8/site-packages/tensorflow/python/eager/function.py", line 1687, in _call_impl return self._call_with_flat_signature(args, kwargs, cancellation_manager) File "/home/chop/Scripts/style-transfer-video-processor/.venv/lib/python3.8/site-packages/tensorflow/python/eager/function.py", line 1736, in _call_with_flat_signature return self._call_flat(args, self.captured_inputs, cancellation_manager) File "/home/chop/Scripts/style-transfer-video-processor/.venv/lib/python3.8/site-packages/tensorflow/python/saved_model/load.py", line 115, in _call_flat return super(_WrapperFunction, self)._call_flat(args, captured_inputs, File "/home/chop/Scripts/style-transfer-video-processor/.venv/lib/python3.8/site-packages/tensorflow/python/eager/function.py", line 1918, in _call_flat return self._build_call_outputs(self._inference_function.call( File "/home/chop/Scripts/style-transfer-video-processor/.venv/lib/python3.8/site-packages/tensorflow/python/eager/function.py", line 555, in call outputs = execute.execute( File "/home/chop/Scripts/style-transfer-video-processor/.venv/lib/python3.8/site-packages/tensorflow/python/eager/execute.py", line 59, in quick_execute tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name, tensorflow.python.framework.errors_impl.NotFoundError: No algorithm worked! [[node InceptionV3/Conv2d_1a_3x3/Conv2D (defined at /home/chop/Scripts/style-transfer-video-processor/.venv/lib/python3.8/site-packages/tensorflow_hub/module_v2.py:106) ]] [Op:__inference_pruned_3217]

Function call stack: pruned `

how to avoid this problem. thanks