Error concealment using direction-oriented candidate set and predicted boundary matching criteria

Chorng Yann Su, Shiou Haur Tsay, Chia Hao Huang

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

6 Citations (Scopus)

Abstract

In this paper, we propose two novel approaches for the error concealment used in H.264 codec. The first one is a direction-oriented candidate set, which decides a proper candidate set of motion vectors by using the trend of collocated motion vector. The second is a predicted boundary matching criterion, which uses one-order prediction of pixel value on boundary to reduce the incurred error of traditional boundary matching criterion. Compared with the bidirectional motion vector refinement (BMVR), the proposed approaches can elevate about 0.75 dB of peak-signal to noise ratio (PSNR) value and perform well especially for the video sequences with moderate and fast motion.

Original languageEnglish
Title of host publication2006 IEEE International Conference on Image Processing, ICIP 2006 - Proceedings
Pages2221-2224
Number of pages4
DOIs
Publication statusPublished - 2006 Dec 1
Externally publishedYes
Event2006 IEEE International Conference on Image Processing, ICIP 2006 - Atlanta, GA, United States
Duration: 2006 Oct 82006 Oct 11

Publication series

NameProceedings - International Conference on Image Processing, ICIP
ISSN (Print)1522-4880

Conference

Conference2006 IEEE International Conference on Image Processing, ICIP 2006
CountryUnited States
CityAtlanta, GA
Period06/10/806/10/11

Keywords

  • Candidate set
  • Error concealment
  • Predicted boundary matching criterion
  • Side matching criterion

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Signal Processing

Fingerprint Dive into the research topics of 'Error concealment using direction-oriented candidate set and predicted boundary matching criteria'. Together they form a unique fingerprint.

Cite this