Change search
ReferencesLink to record
Permanent link

Direct link
Improving Intra Pixel prediction for H.264 video coding Senay
Blekinge Institute of Technology, School of Engineering, Department of Signal Processing.
2008 (English)Independent thesis Advanced level (degree of Master (Two Years))Student thesis
Abstract [en]

H.264 is the latest most efficient video coding standard developed by joint collaboration between ITU-T’s Study Group 16 Question 6 (Video Coding Experts Group VCEG) and ISO’s MPEG forming a group known as the Joint Video Team (JVT). The JVT has added additional features to the previous video coding standards (MPEG-4 and H.263) which has allowed the current H.264 to provide a much more efficient compression ability giving a bit rate reduction of about 50 % of the previous standards for the same video quality. This thesis is among the ongoing studies that deal with trying to improve the intra prediction scheme of H.264. It aims to modify the directional extrapolation scheme of the standard intra prediction and add a combined usage based on the local structure of previously reconstructed pixels surrounding the area to be coded. A new method of searching the direction of the local structure of an area using optimal filters is deployed. The method is based on the least squares algorithm. A modified coder structure is implemented in an attempt to obtain improved coding efficiency using the new prediction. The result based on visual inspection and objective evaluation of the new modified prediction against standard H.264 intra prediction is presented.

Place, publisher, year, edition, pages
2008. , 59 p.
Keyword [en]
H.264, Intra-prediction, wiener filter, local structure
National Category
Signal Processing
URN: urn:nbn:se:bth-4030Local ID: diva2:831349
0736167293Available from: 2015-04-22 Created: 2008-09-08 Last updated: 2015-06-30Bibliographically approved

Open Access in DiVA

fulltext(3061 kB)228 downloads
File information
File name FULLTEXT01.pdfFile size 3061 kBChecksum SHA-512
Type fulltextMimetype application/pdf

By organisation
Department of Signal Processing
Signal Processing

Search outside of DiVA

GoogleGoogle Scholar
Total: 228 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 36 hits
ReferencesLink to record
Permanent link

Direct link