LiteFlowNet3: Resolving Correspondence Ambiguity for More Accurate Optical Flow Estimation

Tak Wai Hui*, Chen Change Loy

*Corresponding author for this work

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

73 Citations (Scopus)

Abstract

Deep learning approaches have achieved great success in addressing the problem of optical flow estimation. The keys to success lie in the use of cost volume and coarse-to-fine flow inference. However, the matching problem becomes ill-posed when partially occluded or homogeneous regions exist in images. This causes a cost volume to contain outliers and affects the flow decoding from it. Besides, the coarse-to-fine flow inference demands an accurate flow initialization. Ambiguous correspondence yields erroneous flow fields and affects the flow inferences in subsequent levels. In this paper, we introduce LiteFlowNet3, a deep network consisting of two specialized modules, to address the above challenges. (1) We ameliorate the issue of outliers in the cost volume by amending each cost vector through an adaptive modulation prior to the flow decoding. (2) We further improve the flow accuracy by exploring local flow consistency. To this end, each inaccurate optical flow is replaced with an accurate one from a nearby position through a novel warping of the flow field. LiteFlowNet3 not only achieves promising results on public benchmarks but also has a small model size and a fast runtime.

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
Pages169-184
Number of pages16
ISBN (Print)9783030585648
DOIs
Publication statusPublished - 2020
Externally publishedYes
Event16th European Conference on Computer Vision, ECCV 2020 - Glasgow, United Kingdom
Duration: Aug 23 2020Aug 28 2020

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume12365 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference16th European Conference on Computer Vision, ECCV 2020
Country/TerritoryUnited Kingdom
CityGlasgow
Period8/23/208/28/20

Bibliographical note

Publisher Copyright:
© 2020, Springer Nature Switzerland AG.

ASJC Scopus Subject Areas

  • Theoretical Computer Science
  • General Computer Science

Cite this