4 March 2013 Fast coding unit selection method for high efficiency video coding intra prediction
Jian Xiong
Author Affiliations +
Abstract
The high efficiency video coding (HEVC) video coding standard under development can achieve higher compression performance than previous standards, such as MPEG-4, H.263, and H.264/AVC. To improve coding performance, a quad-tree coding structure and a robust rate-distortion (RD) optimization technique is used to select an optimum coding mode. Since the RD costs of all possible coding modes are computed to decide an optimum mode, high computational complexity is induced in the encoder. A fast learning-based coding unit (CU) size selection method is presented for HEVC intra prediction. The proposed algorithm is based on theoretical analysis that shows the non-normalized histogram of oriented gradient (n-HOG) can be used to help select CU size. A codebook is constructed offline by clustering n-HOGs of training sequences for each CU size. The optimum size is determined by comparing the n-HOG of the current CU with the learned codebooks. Experimental results show that the CU size selection scheme speeds up intra coding significantly with negligible loss of peak signal-to-noise ratio.
© 2013 Society of Photo-Optical Instrumentation Engineers (SPIE) 0091-3286/2013/$25.00 © 2013 SPIE
Jian Xiong "Fast coding unit selection method for high efficiency video coding intra prediction," Optical Engineering 52(7), 071504 (4 March 2013). https://doi.org/10.1117/1.OE.52.7.071504
Published: 4 March 2013
Lens.org Logo
CITATIONS
Cited by 8 scholarly publications.
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Copper

Video coding

Computer programming

Standards development

Optical engineering

Video

Algorithm development

RELATED CONTENT

4K-based intra and interprediction techniques for HEVC
Proceedings of SPIE (April 29 2016)
DM642 digital media processor
Proceedings of SPIE (May 07 2003)
Benchmarking real-time HEVC streaming
Proceedings of SPIE (May 01 2012)

Back to Top