Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

End-to-End Conditional GAN-based Architectures for Image Colourisation

Abstract: 

In this work recent advances in conditional adversarial networks are investigated to develop an end-to-end architecture based on Convolutional Neural Networks (CNNs) to directly map realistic colours to an input greyscale image. Observing that existing colourisation methods sometimes exhibit a lack of colourfulness, this paper proposes a method to improve colourisation results. In particular, the method uses Generative Adversarial Neural Networks (GANs) and focuses on improvement of training stability to enable better generalisation in large multi-class image datasets. Additionally, the integration of instance and batch normalisation layers in both generator and discriminator is introduced to the popular U-Net architecture, boosting the network capabilities to generalise the style changes of the content. The method has been tested using the ILSVRC 2012 dataset, achieving improved automatic colourisation results compared to other methods based on GANs.

up
0 users have voted:

Paper Details

Authors:
Marc Gorriz Blanch, Marta Mrak, Alan Smeaton, Noel O'Connor
Submitted On:
24 September 2019 - 12:11pm
Short Link:
Type:
Poster
Event:
Presenter's Name:
Marc Gorriz Blanch
Paper Code:
147
Document Year:
2019
Cite

Document Files

MMSP2019_poster.pdf

(21)

Subscribe

[1] Marc Gorriz Blanch, Marta Mrak, Alan Smeaton, Noel O'Connor, "End-to-End Conditional GAN-based Architectures for Image Colourisation", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4834. Accessed: Dec. 10, 2019.
@article{4834-19,
url = {http://sigport.org/4834},
author = {Marc Gorriz Blanch; Marta Mrak; Alan Smeaton; Noel O'Connor },
publisher = {IEEE SigPort},
title = {End-to-End Conditional GAN-based Architectures for Image Colourisation},
year = {2019} }
TY - EJOUR
T1 - End-to-End Conditional GAN-based Architectures for Image Colourisation
AU - Marc Gorriz Blanch; Marta Mrak; Alan Smeaton; Noel O'Connor
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4834
ER -
Marc Gorriz Blanch, Marta Mrak, Alan Smeaton, Noel O'Connor. (2019). End-to-End Conditional GAN-based Architectures for Image Colourisation. IEEE SigPort. http://sigport.org/4834
Marc Gorriz Blanch, Marta Mrak, Alan Smeaton, Noel O'Connor, 2019. End-to-End Conditional GAN-based Architectures for Image Colourisation. Available at: http://sigport.org/4834.
Marc Gorriz Blanch, Marta Mrak, Alan Smeaton, Noel O'Connor. (2019). "End-to-End Conditional GAN-based Architectures for Image Colourisation." Web.
1. Marc Gorriz Blanch, Marta Mrak, Alan Smeaton, Noel O'Connor. End-to-End Conditional GAN-based Architectures for Image Colourisation [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4834