Guided Semantic Flow

Sangryul Jeon, Dongbo Min, Seungryong Kim, Jihwan Choe, Kwanghoon Sohn

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Establishing dense semantic correspondences requires dealing with large geometric variations caused by the unconstrained setting of images. To address such severe matching ambiguities, we introduce a novel approach, called guided semantic flow, based on the key insight that sparse yet reliable matches can effectively capture non-rigid geometric variations, and these confident matches can guide adjacent pixels to have similar solution spaces, reducing the matching ambiguities significantly. We realize this idea with learning-based selection of confident matches from an initial set of all pairwise matching scores and their propagation by a new differentiable upsampling layer based on moving least square concept. We take advantage of the guidance from reliable matches to refine the matching hypotheses through Gaussian parametric model in the subsequent matching pipeline. With the proposed method, state-of-the-art performance is attained on several standard benchmarks for semantic correspondence.

Original languageEnglish
Title of host publicationComputer Vision – ECCV 2020 - 16th European Conference, 2020, Proceedings
EditorsAndrea Vedaldi, Horst Bischof, Thomas Brox, Jan-Michael Frahm
PublisherSpringer Science and Business Media Deutschland GmbH
Pages631-648
Number of pages18
ISBN (Print)9783030586034
DOIs
Publication statusPublished - 2020
Event16th European Conference on Computer Vision, ECCV 2020 - Glasgow, United Kingdom
Duration: 2020 Aug 232020 Aug 28

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume12373 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference16th European Conference on Computer Vision, ECCV 2020
CountryUnited Kingdom
CityGlasgow
Period20/8/2320/8/28

Bibliographical note

Funding Information:
Acknowledgements. This research was supported by Next-Generation Information Computing Development Program through the National Research Foundation of Korea(NRF) funded by the Ministry of Science and ICT (NRF2017M3C4A7069370).

Publisher Copyright:
© 2020, Springer Nature Switzerland AG.

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Guided Semantic Flow'. Together they form a unique fingerprint.

Cite this