Multi-homography Estimation and Inference Driven by Contour Alignment

Tao Cai, Yunde Jia, Huijun Di*, Yuwei Wu

*Corresponding author for this work

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

Abstract

Recently, multiple homography estimation is preferable for image stitching to handle the parallax problem, by estimating homographies from the feature correspondence in each local region. However, correspondence outliers and insufficient feature coverage will lead to unreliable local homography fitting. In this paper, we propose a novel method of multi-homography estimation and inference, driven by contour alignment. Our method uses explicit structural verification through contour alignment to eliminate incorrectly fitted homographies in some regions, and to select a better homography from other regions if current homography is rejected or with worse accuracy. With the guidance of the contour alignment, dense image alignment result is obtained by further inferring the local homography per superpixel. Quantitative and qualitative comparisons demonstrate the effectiveness of our method, especially for scenes with large parallax and viewpoint changes.

Original languageEnglish
Title of host publicationImage and Graphics - 11th International Conference, ICIG 2021, Proceedings
EditorsYuxin Peng, Shi-Min Hu, Moncef Gabbouj, Kun Zhou, Michael Elad, Kun Xu
PublisherSpringer Science and Business Media Deutschland GmbH
Pages539-551
Number of pages13
ISBN (Print)9783030873547
DOIs
Publication statusPublished - 2021
Event11th International Conference on Image and Graphics, ICIG 2021 - Haikou, China
Duration: 6 Aug 20218 Aug 2021

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume12888 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference11th International Conference on Image and Graphics, ICIG 2021
Country/TerritoryChina
CityHaikou
Period6/08/218/08/21

Keywords

  • Contour alignment
  • Homography estimation
  • Image alignment
  • Image stitching

Fingerprint

Dive into the research topics of 'Multi-homography Estimation and Inference Driven by Contour Alignment'. Together they form a unique fingerprint.

Cite this

Cai, T., Jia, Y., Di, H., & Wu, Y. (2021). Multi-homography Estimation and Inference Driven by Contour Alignment. In Y. Peng, S.-M. Hu, M. Gabbouj, K. Zhou, M. Elad, & K. Xu (Eds.), Image and Graphics - 11th International Conference, ICIG 2021, Proceedings (pp. 539-551). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12888 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-87355-4_45