I tried to explain in the README, but there is not a easy way to apply this kind of chroma change, I'm not aware of filters being able to do that.
It could be useful if the Color Panel in vsViewer will be extended to display the values of HSV, currently only RGB and YUV values are shown.
To get the correct range to adjust, at the moment, I save a screenshot of the image with the wrong colors and open it in Paint.NET and use the color picker tool to see the HSV values.
I noted a defect in the current version of Hybrid.
I can add more parameters in the list as shown in this screenshot
But when I uncheck the filters are restored the old values
Traceback (most recent call last):
File "src\cython\vapoursynth.pyx", line 2941, in vapoursynth.publicFunction
File "src\cython\vapoursynth.pyx", line 2943, in vapoursynth.publicFunction
File "src\cython\vapoursynth.pyx", line 683, in vapoursynth.FuncData.__call__
File "F:\Hybrid\64bit\Vapoursynth\Lib\site-packages\torch\utils\_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "F:\Hybrid\64bit\Vapoursynth\Lib\site-packages\vsddcolor\__init__.py", line 94, in inference
output_ab = module(img_gray_rgb)
^^^^^^^^^^^^^^^^^^^^
File "F:\Hybrid\64bit\Vapoursynth\Lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "F:\Hybrid\64bit\Vapoursynth\Lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "F:\Hybrid\64bit\Vapoursynth\Lib\site-packages\vsddcolor\ddcolor_arch.py", line 59, in forward
out_feat = self.decoder()
^^^^^^^^^^^^^^
File "F:\Hybrid\64bit\Vapoursynth\Lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "F:\Hybrid\64bit\Vapoursynth\Lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "F:\Hybrid\64bit\Vapoursynth\Lib\site-packages\vsddcolor\ddcolor_arch.py", line 107, in forward
out0 = self.layers[0](encode_feat)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "F:\Hybrid\64bit\Vapoursynth\Lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "F:\Hybrid\64bit\Vapoursynth\Lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "F:\Hybrid\64bit\Vapoursynth\Lib\site-packages\vsddcolor\ddcolor_arch_utils\unet.py", line 207, in forward
cat_x = self.relu(torch.cat([up_out, self.bn(s)], dim=1))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 24 but got size 25 for tensor number 1 in the list.
Cu Selur
----
Dev versions are in the 'experimental'-folder of my GoogleDrive, which is linked on the download page.
The problem is that the input_size in ddcolor must be divisible by 32. Try to increase the render_factor to 26, and should work.
I will release a fix to ensure that the ddcolor input_size will be divisible by 32.