Semi-supervised semantic matching

Zakaria Laskar*, Juho Kannala

*Corresponding author for this work

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

Abstract

Convolutional neural networks (CNNs) have been successfully applied to solve the problem of correspondence estimation between semantically related images. Due to non-availability of large training datasets, existing methods resort to self-supervised or unsupervised training paradigm. In this paper we propose a semi-supervised learning framework that imposes cyclic consistency constraint on unlabeled image pairs. Together with the supervised loss the proposed model achieves state-of-the-art on a benchmark semantic matching dataset.

Original languageEnglish
Title of host publicationComputer Vision – ECCV 2018 Workshops, Proceedings
EditorsLaura Leal-Taixé, Stefan Roth
Pages444-455
Number of pages12
DOIs
Publication statusPublished - 1 Jan 2019
MoE publication typeA4 Article in a conference publication
EventEuropean Conference on Computer Vision - Munich, Germany
Duration: 8 Sep 201814 Sep 2018
Conference number: 15

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11131 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

ConferenceEuropean Conference on Computer Vision
Abbreviated titleECCV
CountryGermany
CityMunich
Period08/09/201814/09/2018

Keywords

  • Deep-learning
  • Geometric matching
  • Semantic-matching

Fingerprint Dive into the research topics of 'Semi-supervised semantic matching'. Together they form a unique fingerprint.

  • Cite this

    Laskar, Z., & Kannala, J. (2019). Semi-supervised semantic matching. In L. Leal-Taixé, & S. Roth (Eds.), Computer Vision – ECCV 2018 Workshops, Proceedings (pp. 444-455). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11131 LNCS). https://doi.org/10.1007/978-3-030-11015-4_32