DFDNet
DFDNet copied to clipboard
Blind Face Restoration via Deep Multi-scale Component Dictionaries (ECCV 2020)
I see that your code contains only test code and no training code. Can you provide the construction code and dataset of the facial component dictionary.
This face restoration will change the face color and Background color ? is it ?
Thank you for sharing the 512x512 model and related dictionaries. Could you share 256x256 and 128x128 res's at your convenience?
I see that you've already normalized the tensor with the below codes.  But why in the VGG19, you again normalized it with RGB_mean and RGB_std?  And I don't...
is there any way to make these run on gpu? what is the reason behind the fact that they only run on cpu?
```python ############################################################################### ####################### Step 3: Face Restoration ############################## ############################################################################### /usr/local/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and will be removed in 0.15, please use 'weights' instead. warnings.warn( /usr/local/lib/python3.10/site-packages/torchvision/models/_utils.py:223:...
############################################################################### ####################### Step 1: Crop and Align Face ########################### ############################################################################### Crop and Align test1.jpg image ################ No face is detected Crop and Align test2.jpg image ################ No face is detected...
Could not export Python function call 'BlurFunction'. Remove calls to Python functions before export. Did you forget to add @script or @script_method annotation? If this is a nn.ModuleList, add it...