Back to articles
Articles
Volume: 31 | Article ID: art00013
Image
Parameter optimization in H.265 rate-distortion by single frame semantic scene analysis
Ahmed M Hamza
Abdelrahman Abdelazim
Djamel Ait-Boudaoud
  DOI :  10.2352/ISSN.2470-1173.2019.11.IPAS-262  Published Online :  January 2019
Abstract

The H.265/HEVC (High Efficiency Video Coding) codec and its 3D extensions have crucial rate-distortion mechanisms that help determine coding efficiency. We have introduced in this work a new system of Lagrangian parameterization in RDO cost functions, based on semantic cues in an image, starting with the current HEVC formulation of the Lagrangian hyper-parameter heuristics. Two semantic scenery flag algorithms are presented and tested within the Lagrangian formulation as weighted factors. The investigation of whether the semantic gap between the coder and the image content is holding back the block-coding mechanisms as a whole from achieving greater efficiency has yielded a positive answer.

Journal Title : Electronic Imaging
Publisher Name : Society for Imaging Science and Technology
Publisher Location : 7003 Kilworth Lane, Springfield, VA 22151 USA
Subject Areas :
Views 25
Downloads 2
 DOI 10.2352/ISSN.2470-1173.2019.11.IPAS-262
 articleview.views 25
 articleview.downloads 2
  Cite this article 

Ahmed M Hamza, Abdelrahman Abdelazim, Djamel Ait-Boudaoud, "Parameter optimization in H.265 rate-distortion by single frame semantic scene analysis"  in Proc. IS&T Int’l. Symp. on Electronic Imaging: Image Processing: Algorithms and Systems XVII,  2019,  pp 262-1 - 262-6,  https://doi.org/10.2352/ISSN.2470-1173.2019.11.IPAS-262

 Copy citation
  Copyright statement 
Copyright © Society for Imaging Science and Technology 2019
72010604
Electronic Imaging
2470-1173
Society for Imaging Science and Technology
7003 Kilworth Lane, Springfield, VA 22151 USA
10.2352/ISSN.2470-1173.2019.11.IPAS-262
2470-1173(20190113)2019:11L.2621;1-
ei_24701173_v2019n11_input/s12.xml
/ist/ei/2019/00002019/00000011/art00013
Articles
Parameter optimization in H.265 rate-distortion by single frame semantic scene analysis
HamzaAhmed M
AbdelazimAbdelrahman
Ait-BoudaoudDjamel
13012019
2019
11
262-1
262-6
2019
The H.265/HEVC (High Efficiency Video Coding) codec and its 3D extensions have crucial rate-distortion mechanisms that help determine coding efficiency. We have introduced in this work a new system of Lagrangian parameterization in RDO cost functions, based on semantic cues in an image, starting with the current HEVC formulation of the Lagrangian hyper-parameter heuristics. Two semantic scenery flag algorithms are presented and tested within the Lagrangian formulation as weighted factors. The investigation of whether the semantic gap between the coder and the image content is holding back the block-coding mechanisms as a whole from achieving greater efficiency has yielded a positive answer.
Video CodingHEVCoptimization