Correction of geometric perceptual distortions in pictures

Denis Zorin, Alan H. Barr

Research output: Contribution to conferencePaperpeer-review

Abstract

An approach based on a mathematical formalization of desirable properties of pictures is presented for correcting several types of perceived geometric distortions in computer-generated and photographic images. Perceptually preferable viewing transformations are obtained from a small set of simple assumptions and shown to be decomposed into a perspective or parallel projection followed by a planar transformation. Following this, it is proven that two perceptually important properties are incompatible and cannot be satisfied simultaneously. By implementing the proposed transformations to computer-generated and photographic perspective projection images, it is found that the transformations considerably reduced distortion in wide-angle motion pictures and computer-generated animations.

Original languageEnglish (US)
Pages257-264
Number of pages8
DOIs
StatePublished - 1995
EventProceedings of the 22nd Annual ACM Conference on Computer Graphics and Interactive Techniques - Los Angeles, CA, USA
Duration: Aug 9 1995Aug 11 1995

Other

OtherProceedings of the 22nd Annual ACM Conference on Computer Graphics and Interactive Techniques
CityLos Angeles, CA, USA
Period8/9/958/11/95

ASJC Scopus subject areas

  • Computer Science(all)

Fingerprint Dive into the research topics of 'Correction of geometric perceptual distortions in pictures'. Together they form a unique fingerprint.

Cite this