Regularization of Building Boundaries in Satellite Images Using Adversarial and Regularized Losses

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Abstract

In this paper we present a method for building boundary re-finement and regularization in satellite images using a fullyconvolutional neural network trained with a combination ofadversarial and regularized losses. Compared to a pure MaskR-CNN model, the overall algorithm can achieve equivalentperformance in terms of accuracy and completeness. How-ever, unlike Mask R-CNN that produces irregular footprints,our framework generates regularized and visually pleasingbuilding boundaries which are beneficial in many applica-tions.
Original languageEnglish
Title of host publicationIGARSS 2019
PublisherIEEE Publications
Pages5140-5143
Number of pages4
Publication statusPublished - 1 Jul 2019

Keywords

  • Generative adversarial networks
  • build-ing segmentation
  • boundary refinement
  • satellite images

Cite this

Regularization of Building Boundaries in Satellite Images Using Adversarial and Regularized Losses. / Zorzi, Stefano; Fraundorfer, Friedrich.

IGARSS 2019. IEEE Publications, 2019. p. 5140-5143.

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

@inproceedings{8317a89ea4ff4e0d9994432e6ae0287c,
title = "Regularization of Building Boundaries in Satellite Images Using Adversarial and Regularized Losses",
abstract = "In this paper we present a method for building boundary re-finement and regularization in satellite images using a fullyconvolutional neural network trained with a combination ofadversarial and regularized losses. Compared to a pure MaskR-CNN model, the overall algorithm can achieve equivalentperformance in terms of accuracy and completeness. How-ever, unlike Mask R-CNN that produces irregular footprints,our framework generates regularized and visually pleasingbuilding boundaries which are beneficial in many applica-tions.",
keywords = "Generative adversarial networks, build-ing segmentation, boundary refinement, satellite images",
author = "Stefano Zorzi and Friedrich Fraundorfer",
year = "2019",
month = "7",
day = "1",
language = "English",
pages = "5140--5143",
booktitle = "IGARSS 2019",
publisher = "IEEE Publications",

}

TY - GEN

T1 - Regularization of Building Boundaries in Satellite Images Using Adversarial and Regularized Losses

AU - Zorzi, Stefano

AU - Fraundorfer, Friedrich

PY - 2019/7/1

Y1 - 2019/7/1

N2 - In this paper we present a method for building boundary re-finement and regularization in satellite images using a fullyconvolutional neural network trained with a combination ofadversarial and regularized losses. Compared to a pure MaskR-CNN model, the overall algorithm can achieve equivalentperformance in terms of accuracy and completeness. How-ever, unlike Mask R-CNN that produces irregular footprints,our framework generates regularized and visually pleasingbuilding boundaries which are beneficial in many applica-tions.

AB - In this paper we present a method for building boundary re-finement and regularization in satellite images using a fullyconvolutional neural network trained with a combination ofadversarial and regularized losses. Compared to a pure MaskR-CNN model, the overall algorithm can achieve equivalentperformance in terms of accuracy and completeness. How-ever, unlike Mask R-CNN that produces irregular footprints,our framework generates regularized and visually pleasingbuilding boundaries which are beneficial in many applica-tions.

KW - Generative adversarial networks

KW - build-ing segmentation

KW - boundary refinement

KW - satellite images

M3 - Conference contribution

SP - 5140

EP - 5143

BT - IGARSS 2019

PB - IEEE Publications

ER -