Detection method and removing filter for corner outliers in highly compressed video

  • Kim, Jongho
  • Jeon, Kicheol
  • Jeong, Jechang
Citations

SCOPUS

0

초록

We propose a detection method and a removal filter for corner outliers in order to improve visual quality in highly compressed video. Corner outliers are detected using the direction of edge going through a block-corner and the properties of blocks around the block-corner. The proposed filter for removing corner outliers, which compensates the stair-shaped discontinuities around edges using the adjacent pixels, is applied to the detected area. Simulation results show that the proposed method improves, particularly in combination with deblocking filters, the visual quality remarkably.

키워드

Block-based codingCorner outlierLow bit-rate videoMPEG-4 videoImage codingImage compressionMotion Picture Experts Group standardsStatisticsBlock basedCompressed videoCorner outlierDeblocking filtersDetection methodsLow Bit RateMPEG-4 videoVisual qualitiesEdge detection
제목
Detection method and removing filter for corner outliers in highly compressed video
저자
Kim, JonghoJeon, KicheolJeong, Jechang
DOI
10.1007/11949534_119
발행일
2006-12
유형
Conference Paper
저널명
Lecture Notes in Computer Science
4319 LNCS
페이지
1177 ~ 1184