Compression of indoor video sequences using homography-based segmentation

Tae Joon Park, S. Fleishman, D. Cohen-Or, D. Lischinski

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

1 Scopus citations

Abstract

We present a new compression algorithm for video sequences of indoor scenes, or more generally, sequences containing mostly planar and near-planar surfaces. Our approach utilizes edge and optical flow information in order to segment selected keyframes into regions of general shape, such that the motion of each region is predicted well by a planar homography. With this kind of motion prediction, the errors between the predicted intermediate frames and the actual ones are very small, and can be compactly encoded. Our results demonstrate significant improvements in the accuracy of the compressed video sequences, compared to standard general purpose video compression.

Original languageAmerican English
Title of host publicationProceedings - 8th Pacific Conference on Computer Graphics and Applications, PG 2000
EditorsBrian A. Barsky, Yoshihisa Shinagawa, Wenping Wang
PublisherIEEE Computer Society
Pages290-299 and 450
ISBN (Electronic)0769508685
DOIs
StatePublished - 2000
Event8th Pacific Conference on Computer Graphics and Applications, PG 2000 - Hong Kong, China
Duration: 3 Oct 20005 Oct 2000

Publication series

NameProceedings - Pacific Conference on Computer Graphics and Applications
Volume2000-January
ISSN (Print)1550-4085

Conference

Conference8th Pacific Conference on Computer Graphics and Applications, PG 2000
Country/TerritoryChina
CityHong Kong
Period3/10/005/10/00

Bibliographical note

Publisher Copyright:
© 2000 IEEE.

Keywords

  • Broadcasting
  • Computer science
  • Image coding
  • Image storage
  • Layout
  • MPEG 4 Standard
  • Multimedia communication
  • Transform coding
  • Video compression
  • Video sequences

Fingerprint

Dive into the research topics of 'Compression of indoor video sequences using homography-based segmentation'. Together they form a unique fingerprint.

Cite this