Skip to main navigation Skip to search Skip to main content

Reconstruction of 3D object space from imagery by feature-based matching

  • Ohio State University

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

2 Scopus citations

Abstract

the visible surface of the 3D object space can be explicitly described by smooth surface patches and their boundaries. The surface patches are approximated by smooth functions and the fitting error characterizes the roughness. The boundaries of smooth surface patches are characterized by rapid changes in elevations or surface normals. For reconstructing surfaces automatically it is desirable to determine the boundaries explicitly, in a direct way. Imagery offers this possibility because boundaries in object space generally appear as edges. The main thrust of the paper in on matching features from images. Every matched edge is intersected in object space. Matching of conjugate edges is performed on back-projection and Iterative Closest Point algorithm. This paper concludes with sample test and considerations

Original languageEnglish
Title of host publication25th Anniversary IGARSS 2005
Subtitle of host publicationIEEE International Geoscience and Remote Sensing Symposium
Pages3967-3970
Number of pages4
DOIs
StatePublished - 2005
Event2005 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2005 - Seoul, Korea, Republic of
Duration: Jul 25 2005Jul 29 2005

Publication series

NameInternational Geoscience and Remote Sensing Symposium (IGARSS)
Volume6

Conference

Conference2005 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2005
Country/TerritoryKorea, Republic of
CitySeoul
Period07/25/0507/29/05

Keywords

  • 3D object space
  • Back-projection
  • Component
  • Feature matching
  • Iterative Closest Point

Fingerprint

Dive into the research topics of 'Reconstruction of 3D object space from imagery by feature-based matching'. Together they form a unique fingerprint.

Cite this