Gimp and neural networks

Discoloration

The result leaves room for manual rework. For example, the trouser leg is missing color. The defects take us nicely to the subject of the limitations of this approach. For images with many different colors, colorization does not work, or it only works with restrictions. The software tries to calculate a mean value and assumes that the colors are similar.

Even CNN cannot guess colors. If it discovers a currently unclassified object in a black-and-white photo, say a tent [11], it guesses its color on the basis of the dominant colors – and can get things totally wrong depending on the circumstances. Having said this, the model achieves "natural-looking" [4] results, according to the study.

Another useful deep-learning project in the image-processing field is the neural network by the Chinese University of Hong Kong, which knocks out people from photos [12], eliminating the background and leaving the human image only. The procedure demonstrated in the paper is not perfect, but it does achieve better results than the features built into current image-editing programs. The Chinese tool also uses a CNN.

Infos

  1. SIGGRAPH 2016: http://s2016.siggraph.org
  2. Github repository for the project: https://github.com/satoshiiizuka/siggraph2016_colorization
  3. Project website: http://hi.cs.waseda.ac.jp/~iizuka/projects/colorization/en/
  4. Scientific work on the project: http://hi.cs.waseda.ac.jp/~iizuka/projects/colorization/data/colorization_sig2016.pdf
  5. Convolutional neural networks: https://annalyzin.wordpress.com/2016/01/26/introduction-to-convolutional-neural-network/
  6. CNN classifications: http://cs231n.github.io/classification/
  7. Torch: http://torch.ch
  8. Lua: https://www.lua.org
  9. Lab color model: https://docs.gimp.org/de/glossary.html#glossary-lab
  10. Gimp's decompose plug-in: https://docs.gimp.org/de/plug-in-decompose-registered.html
  11. Project presentation: http://hi.cs.waseda.ac.jp/~iizuka/projects/colorization/data/IizukaSIGGRAPH2016_slide.pdf
  12. Knocking out images with deep learning: http://www.cse.cuhk.edu.hk/leojia/papers/portrait_eg16.pdf

« Previous 1 2

Buy this article as PDF

Express-Checkout as PDF
Price $2.95
(incl. VAT)

Buy Linux Magazine

SINGLE ISSUES
Print Issues
Digital Issues
 
SUBSCRIPTIONS
Print Subs
Digisubs
 
TABLET & SMARTPHONE APPS
Get it on Google Play

US / Canada

Get it on Google Play

UK / Australia

Related content

  • FAQ

    Welcome our new artificial intelligence overlords by tinkering with their gray matter.

    more »
  • Spam-Detecting Neural Network

    Build a neural network that uncovers spam websites.

    more »
  • Neural Networks

    3, 4, 8, 11… ? A neural network can complete this series without knowledge of the underlying algorithm – by a kind of virtual gut feeling. We’ll show you how neural networks solve problems by simulating the behavior of a human brain.

    more »
  • Animation with GIMP

    Although developed for editing individual images, GIMP has everything you need to create perfect animations via plugins and scripts.

    more »
  • Tiling in GIMP

    Graphic artists often face the problem of turning a photograph into an image that will tile over a larger surface. This task is not as easy as it sounds, but if you’re up for the challenge, this tutorial will give you a first-hand look at some advanced tools in the GIMP toolkit.

    more »
comments powered by Disqus

Visit Our Shop

Trial Subscription

Digital Subscription

Subscribe

Direct Download

Read full article as PDF:

Price $2.95

News

Tag Cloud

Administration Community Desktop Events Hardware Linux Linux Pro Magazine Mobile Programming Software Ubuntu Web Development Windows free software open source