A Parallel Network for the Computation of Structure from Long-Range Motion

Robert Laganière, Frédéric Labrosse, P. Cohen

Research output: Chapter in Book/Report/Conference proceedingConference Proceeding (Non-Journal item)

Abstract

We propose a parallel architecture to compute the 3-D structure of a moving scene from a long image sequence, using a principle known as the Incremental Rigidity Scheme. At each instant an internal model of the 3-D structure is updated, based upon the observations accumulated until now. The updating process favors rigid transformations but tolerates a limited deviation from rigidity. This deviation eventually leads the internal model to converge towards the actual 3-D structure of the scene. The main advantage of this architecture is its ability to accurately estimate the 3-D structure of the scene, at a low computational cost. Testing has been succesfully performed on synthetic data as well as real image sequences.

Original languageEnglish
Title of host publicationProceedings - 1992 International Joint Conference on Neural Networks, IJCNN 1992
PublisherIEEE Press
Pages278-285
Number of pages8
ISBN (Electronic)0780305590
ISBN (Print)0-7803-0559-0
DOIs
Publication statusPublished - 1992
Externally publishedYes
Event1992 International Joint Conference on Neural Networks, IJCNN 1992 - Baltimore, United States of America
Duration: 07 Jun 199211 Jun 1992

Publication series

NameProceedings of the International Joint Conference on Neural Networks
Volume3

Conference

Conference1992 International Joint Conference on Neural Networks, IJCNN 1992
Country/TerritoryUnited States of America
CityBaltimore
Period07 Jun 199211 Jun 1992

Keywords

  • computer networks
  • concurrent computing
  • layout
  • image sequences
  • parallel architectures
  • image converters
  • computer architecture
  • computational efficiency
  • testing
  • performance evaluation

Fingerprint

Dive into the research topics of 'A Parallel Network for the Computation of Structure from Long-Range Motion'. Together they form a unique fingerprint.

Cite this