Skip to content

RaymondDashWu/style-transfer-app

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

14 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Note: npy file in vgg_conv.npy.zip in case the box link ever goes down.

Tried deploying to Lambda via Zappa and Serverless. File was too big but when I finally got it under 512MB on Zappa I had to set "use_precompiled_packages": false which disabled C extensions. This affected _imaging from Pillow and

TODO

  • README
  • Google bucket to upload to Imgur Note: couldn't figure out how to upload file objects to Imgur Figured out
  • Imgur create gallery to store input
  • Create frontend rather than Flask restplus
  • Style all the pages
  • Feedback form with star rating and description
  • Separate pages explaining how it works, link to repo, etc
  • Output
  • Hook into React gallery
  • Calculate estimated time based on tf loss
  • Email user when output is finished
  • Make separate Imgur account
  • Hook up Imgur account to app
  • Check to see if requirements.txt works. Used pipreqs to get a cleaner file
  • Account for bad input
  • Make Imgur images public when uploaded

Below is Vince's original README explaining how this version of style transfer works


Style Transfer as Optimal Transport

An algorithm that transfers the distribution of visual characteristics, or style, of a reference image onto a subject image via an Optimal Transport plan.

input-content

How it Works:

tl;dr - Passes a subject and style image through the convolution layers of the vgg19 network. Extracts probabalistic descriptions (first two moments) of the convolution filter activations. Calculates L2-Wasserstein distance between these probability distributions and then modifies the subject image optimally to minimize this distance.

Running

Requires:

  1. Python 3 (w/ NumPy and PIL)
  2. Tensorflow (tested w/ version 1.3)
  3. 'vgg_conv.npy' binary (115.5 MB) which contains calibrated convolution filters from vgg19 network. Can be downloaded here [md5sum: bf8a930fec201a0a2ade13d3f7274d0e]

Basic Usage from Command Line:

python basic_styletrans.py --subject media/wave_small.jpg --style media/kngwa_small.jpg --output media/wave_kngwa.jpg --vggnet vgg_conv.npy

Output:

loss: 118,500,040.00
ALL DONE 
post clip | time: 49.0 final loss: 2,771,486.50
synthesized image saved: media/wave_kngwa.jpg

input-content

About

A deployed Flask version of Vince Marron's excellent style transfer work https://github.com/VinceMarron/style_transfer

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published