Skip to content

An implementation of "Perceptual Losses for Real-Time Style Transferand Super-Resolution" in TensorFlow 1.14

Notifications You must be signed in to change notification settings

qbeer/fast-neural-style-transfer

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

34 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Fast neural style transfer

  • With VGG16 loss head.
  • With UNET fully convolutional net for image generation.

From style images:

Van Gogh Picasso

To train the network with Imagenette:

pipenv shell
pipenv install
pipenv run python train.py --starry_night --width=<width_int> --height=<height-int>

Examples

  • Heidelberg - images
Original Van Gogh Picasso

Videos

  • Outside scene with friends
Van Gogh Picasso
mount-vesuvio mount-vesuvio
  • Mount Vesuvio's inside
Van Gogh Picasso
outside-scene-with-friends outside-scene-with-friends

References

@Alex

About

An implementation of "Perceptual Losses for Real-Time Style Transferand Super-Resolution" in TensorFlow 1.14

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages