-
I have 2 1080ti's. When I use "-g 2" both cards have utilization that bounces between 30-60%. Using just one I get almost 100% utilization. It also seems that training is about 2 to 3 times faster whe…
-
I am trying to use GAN support in my faceswaps as my previous attempt of 3 days of training yeilded terrible results with 3k pictures (a good blend) for each target. I see that the python scripts are …
-
INFO Setup in Windows 10
INFO Installed Python: 3.6.7 64bit
INFO Upgrading pip...
INFO Installed pip: 18.1
Enable Docker? [y/N] n
INFO Docker Disabled
Enable CUDA? [Y/n] y
INFO…
-
already did subj.
![out00002_0](https://user-images.githubusercontent.com/8076202/37877447-8aac2a2c-306c-11e8-8bc4-e804ff7af6d0.png)
but I will not release it for a while, because I am making…
-
```py
self.erosion_kernel = None
self.blurSize = 8 # Default: 2
self.seamlessClone = True # Default: False
self.maskType = 'FaceHullAndRect' # Default: Rect
```
In my testing, this produces far …
-
Hi,
while reading the code of the decoder I noticed that you are using PixelShuffler instead of Transposed Conv to increase the spatial dimension of the output. This is done by mapping some of the 'c…
-
First check this out, original paper included. https://github.com/titu1994/Image-Super-Resolution
One of the big problems with faceswaping is close-ups because generated faces are 64x64 -if it's no…
-
I'm working on Windows 10.
I've installed Cuda 9, CuDNN 7 and TensorFlow 1.5, and everything has been working fine for a week now.
I think it's a bit faster, but I have not measured it.
Is there an…
-
## Expected behavior
When performing extraction using the following switches, facial images and alignment data are extracted for each image.
```python Libraries/faceswap/faceswap.py extract -i /…
-
I try using python3 faceswap.py -h in terminal
And I keep getting:
>
Call this program like this:
./face_recognition.py shape_predictor_5_face_landmarks.dat dlib_face_recognition_resnet_mode…