Variational model for depth estimation from images

Alexander Malyshev, Xue-Cheng TAI

Research output: Chapter in book/report/conference proceedingConference proceedingpeer-review

Abstract

We propose a variant of convex reformulation of the standard variational model with non-convex data terms. The proposed convex relaxation of multilabel problems is a continuous formulation of Ishikawa's method for extension of the graph min-cut to the multilabel problems. Our convex continuous reformulation is based upon functional lifting to a higher-dimensional space using superlevel functions. We solve the resulting convex variational problem by the augmented Lagrangian method. The most time consuming part of this method is the numerical solution of a boundary value problem for the Poisson equation in three-dimensional space, which is implemented by means of a fast Poisson solver. We illustrate the developed theory with several numerical examples for the standard correspondence problem for a rectified stereo image pair.

Original languageEnglish
Title of host publication2019 13th International Conference on Software, Knowledge, Information Management and Applications, SKIMA 2019
PublisherIEEE
ISBN (Electronic)9781728127415
DOIs
Publication statusPublished - Aug 2019
Event13th International Conference on Software, Knowledge, Information Management and Applications, SKIMA 2019 - Island of Ulkulhas, Maldives
Duration: 26 Aug 201928 Aug 2019

Publication series

Name2019 13th International Conference on Software, Knowledge, Information Management and Applications, SKIMA 2019

Conference

Conference13th International Conference on Software, Knowledge, Information Management and Applications, SKIMA 2019
Country/TerritoryMaldives
CityIsland of Ulkulhas
Period26/08/1928/08/19

Scopus Subject Areas

  • Artificial Intelligence
  • Computer Science Applications
  • Software
  • Information Systems and Management

User-Defined Keywords

  • Computer vision
  • Minimization methods
  • Optimization

Fingerprint

Dive into the research topics of 'Variational model for depth estimation from images'. Together they form a unique fingerprint.

Cite this