Skip to content

Tensorflow code for paper: Deformable Generator Network: Unsupervised Disentanglement of Appearance and Geometry

Notifications You must be signed in to change notification settings

ruiqigao/DeformableGenerator

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Train the model

  1. Put the folder of training images into ./data. E.g. a folder 'face'
  2. Train the model: python main.py --category face

Test the model

python main.py --category face --test True --ckpt model.ckpt-990 --sample_size 100

Output

***_app.png: generated appearance (without landmarks warping)

***_img.png: generated image (appearance + landmarks)

About

Tensorflow code for paper: Deformable Generator Network: Unsupervised Disentanglement of Appearance and Geometry

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages