Adaptive bidirectional time-recursive interpolation for deinterlacing

Jelena Kovacevic, Robert J. Safranek, Edmund M. Yeh

Research output: Contribution to journalConference article

Abstract

A newly developed algorithm for deinterlacing is presented. It interpolates the missing pixels using a weighted combination of spatial and temporal methods. The algorithm is self-adaptive since it weights various processing blocks based on the error they introduce. It is embedded; it vertically interpolates first, followed by temporal interpolation. The result is used to orientationally interpolate the field which in turn is used to perform motion compensated interpolation. Motion compensation is performed bidirectionally to allow for better performance in case of scene changes, or covered objects. Experiments were run on both real world and computer generated video sequences and results were compared to the original.

Original languageEnglish (US)
Number of pages1
JournalData Compression Conference Proceedings
StatePublished - Jan 1 1995
EventProceedings of the 5th Data Compression Conference - Snowbird, UT, USA
Duration: Mar 28 1995Mar 30 1995

ASJC Scopus subject areas

  • Computer Networks and Communications

Fingerprint Dive into the research topics of 'Adaptive bidirectional time-recursive interpolation for deinterlacing'. Together they form a unique fingerprint.

  • Cite this