Recognising rotationally symmetric surfaces from their outlines

David A. Forsyth, Joseph L. Mundy, Andrew Zisserman, Charles A. Rothwell

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


Recognising a curved surface from its outline in a single view is a major open problem in computer vision. This paper shows techniques for recognising a significant class of surfaces from a single perspective view. The approach uses geometrical facts about bitangencies, creases, and inflections to compute descriptions of the surface’s shape from its image outline. These descriptions are unaffected by the viewpoint or the camera parameters. We show, using images of real scenes, that these representations identify surfaces from their outline alone. This leads to fast and effective recognition of curved surfaces. The techniques we describe work for surfaces that have a rotational symmetry, or are projectively equivalent to a surface with a rotational symmetry, and can be extended to an even larger class of surfaces. All the results in this paper are for the case of full perspective. The results additionally yield techniques for identifying the line in the image plane corresponding to the axis of a rotationally symmetric surface, and for telling whether a surface is rotationally symmetric or not from its outline alone.

Original languageEnglish (US)
Title of host publicationComputer Vision - ECCV 1992 - 2nd European Conference on Computer Vision, Proceedings
EditorsGiulio Sandini
Number of pages9
ISBN (Print)9783540554264
StatePublished - 1992
Externally publishedYes
Event2nd European Conference on Computer Vision, ECCV 1992 - Santa Margherita Ligure, Italy
Duration: May 19 1992May 22 1992

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume588 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Other2nd European Conference on Computer Vision, ECCV 1992
CitySanta Margherita Ligure

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)


Dive into the research topics of 'Recognising rotationally symmetric surfaces from their outlines'. Together they form a unique fingerprint.

Cite this