Repository logo

Infoscience

  • English
  • French
Log In
Logo EPFL, École polytechnique fédérale de Lausanne

Infoscience

  • English
  • French
Log In
  1. Home
  2. Academic and Research Output
  3. Conferences, Workshops, Symposiums, and Seminars
  4. TopoAL: An Adversarial Learning Approach for Topology-Aware Road Segmentation
 
conference paper

TopoAL: An Adversarial Learning Approach for Topology-Aware Road Segmentation

Vasu, Subeesh  
•
Kozinski, Mateusz  
•
Citraro, Leonardo  
Show more
August 23, 2020
[Proceedings of ECCV '20]
European Conference on Computer Vision (ECCV)

Most state-of-the-art approaches to road extraction from aerial images rely on a CNN trained to label road pixels as foreground and remainder of the image as background. The CNN is usually trained by minimizing pixel-wise losses, which is less than ideal to produce binary masks that preserve the road network's global connectivity. To address this issue, we introduce an Adversarial Learning (AL) strategy tailored for our purposes. A naive one would treat the segmentation network as a generator and would feed its output along with ground-truth segmentations to a discriminator. It would then train the generator and discriminator jointly. We will show that this is not enough because it does not capture the fact that most errors are local and need to be treated as such. Instead, we use a more sophisticated discriminator that returns a label pyramid describing what portions of the road network are correct at several different scales. This discriminator and the structured labels it returns are what gives our approach its edge and we will show that it outperforms state-of-the-art ones on the challenging RoadTracer dataset.

  • Files
  • Details
  • Metrics
Type
conference paper
Author(s)
Vasu, Subeesh  
Kozinski, Mateusz  
Citraro, Leonardo  
Fua, Pascal  
Date Issued

2020-08-23

Published in
[Proceedings of ECCV '20]
Total of pages

17

Subjects

Road networks

•

Adversarial learning

•

Generative adversarial network

•

Topology learning

Editorial or Peer reviewed

REVIEWED

Written at

EPFL

EPFL units
CVLAB  
Event nameEvent placeEvent date
European Conference on Computer Vision (ECCV)

[Online event]

August 23-28, 2020

Available on Infoscience
July 22, 2020
Use this identifier to reference this record
https://infoscience.epfl.ch/handle/20.500.14299/170279
Logo EPFL, École polytechnique fédérale de Lausanne
  • Contact
  • infoscience@epfl.ch

  • Follow us on Facebook
  • Follow us on Instagram
  • Follow us on LinkedIn
  • Follow us on X
  • Follow us on Youtube
AccessibilityLegal noticePrivacy policyCookie settingsEnd User AgreementGet helpFeedback

Infoscience is a service managed and provided by the Library and IT Services of EPFL. © EPFL, tous droits réservés