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
- SIGGRAPH 2016: http://s2016.siggraph.org
- Github repository for the project: https://github.com/satoshiiizuka/siggraph2016_colorization
- Project website: http://hi.cs.waseda.ac.jp/~iizuka/projects/colorization/en/
- Scientific work on the project: http://hi.cs.waseda.ac.jp/~iizuka/projects/colorization/data/colorization_sig2016.pdf
- Convolutional neural networks: https://annalyzin.wordpress.com/2016/01/26/introduction-to-convolutional-neural-network/
- CNN classifications: http://cs231n.github.io/classification/
- Torch: http://torch.ch
- Lua: https://www.lua.org
- Lab color model: https://docs.gimp.org/de/glossary.html#glossary-lab
- Gimp's decompose plug-in: https://docs.gimp.org/de/plug-in-decompose-registered.html
- Project presentation: http://hi.cs.waseda.ac.jp/~iizuka/projects/colorization/data/IizukaSIGGRAPH2016_slide.pdf
- 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
(incl. VAT)
Buy Linux Magazine
US / Canada
UK / Australia
Direct Download
Read full article as PDF:
Price $2.95
News
-
NixOS 21.11 Now Available for Download
NixOS “Porcupine” has been made available for installation and includes numerous improvements.
-
KDE Plasma Developers Introduce a GNOME-Like Overview
A new overview effect makes it easier for users to interact with both Plasma Activities and Virtual Desktops.
-
Rocky Linux 8.5 Now Available with Secure Boot Support
The latest iteration of CentOS alternative, Rocky Linux has arrived and includes numerous updates as well as support for Secure Boot.
-
System76 Developing a New Desktop Environment
The makers of Pop!_OS are currently in the early stages of creating a brand new, non-GNOME, desktop environment.
-
Hetzner Opens New Location in the USA
The Hetzner cost-effective Cloud solution has arrived in the United States to offer cloud servers that offer solid performance without a high price.
-
KDE Plasma 5.24 Introduces Fingerprint Reader Support
The next release of the KDE Plasma desktop environment will include support for fingerprint readers for session authentication and more.
-
Ubuntu 21.10 Released and Finally Includes Gnome 40
The most anticipated update to Ubuntu is finally here and the workflow couldn't be better.
-
Linux Can Now Run on Apple’s M1 Chipset
Linux users looking to take advantage of Apple Silicon are about to get their wish with the M1 chipset.
-
MX Linux 21 RC Candidate Available For Testing
The MX Linux development team has released the RC images for MX Linux 21 for testing purposes.
-
Fedora 35 is Shaping Up to Be an Impressive Release
Although the beta version of Fedora’s upcoming release is behind schedule, the Linux distribution is still slotted for October and will include numerous bug fixes and some exciting new features.