Adaptive low-complexity motion estimation
algorithm for high efficiency video coding
encoder
ISSN 1751-9659
Received on 10th June 2015
Revised on 21st December 2015
Accepted on 1st January 2016
doi: 10.1049/iet-ipr.2015.0666
www.ietdl.org
Ahmed Medhat
1
✉
, Ahmed Shalaby
1
, Mohammed Sharaf Sayed
1,2
, Maha Elsabrouty
1
,
Farhad Mehdipour
3
1
ECE Department, Egypt-Japan University of Science and Technology (E-JUST), Alexandria, Egypt
2
ECE Department, Zagazig University, Zagazig, Egypt
3
E-JUST Center, Kyushu University and WaioMio Ltd., New Zealand
✉ E-mail: ahmed.abdelsalam@ejust.edu.eg
Abstract: High quality videos became an essential requirement in recent applications. High efficiency video coding (HEVC)
standard provides an efficient solution for high quality videos at lower bit rates. On the other hand, HEVC comes with
much higher computational cost. In particular, m otion estimatio n ( ME) in HEVC , consumes the largest amount of
computations. Therefore, fast ME algorit hms and hardware accelerators are proposed in order to speed-up integer ME
in HEVC. This study presents a fast centre search algorithm (FCSA) and an adaptive search window algorit hm (ASWA)
for integer pixel ME in HEVC. In addition, centre adaptive search algorithm, a combination of the two proposed
algorithms FCSA and ASWA, is proposed in order to achieve the best performance. Experimental results show notable
speed-upintermsofencodingtimeandbitratesavingwith tolerable peak signal-to-noise ratio (PSNR) quality
degradation. The proposed fast search algorithms reduce the computational complexity of the HEVC encoder by 57%.
This i mproveme nt is accompanied with a modest average PSNR loss of 0.014 dB and an i ncrease by 0.6385% in terms
of bit rate when compared with related works.
1 Introduction
Recently, real-time videos consume more than 50% of current
network traffic, and this share is expected to rise up to 90% within
a few years [1]. In addition, the rising demand for high definition
(HD) and beyond HD videos as well mandates better video
compression efficiency. These facts and needs raise the demand
for new video coding standard with high compression efficiency
compared with the currently used H.264/MPEG-4 advanced video
coding (AVC) standard. The new high efficiency video coding
(HEVC) standard was introduced targeting to double the
compression efficiency with respect to the previous standard
H.264. It can achieve 50% bit rate saving compared with H.264/
MPEG-4 AVC for the same video quality [2, 3].
Quad-tree structure is a basic feature that distinguishes HEVC
from MPEG-4 AVC. As shown in Fig. 1, HEVC is based on code
tree unit (CTU) instead of the macroblock in H.264/MPEG-4
AVC. The size of CTU is variable, unlike traditional macroblock.
CTU size is selected by the encoder and can be larger or smaller
than the traditional macroblock size [2]. Each CTU is split into
one or more code unit (CU), and each CU is further partitioned
into prediction units and a tree of transform units [4].
Motion estimation (ME) is used for inter frame prediction where
the motion in the current frame is predicted from one or more
reference frames. ME entails the major computation complexity
load and processing time in video encoder. In most
implementations, it takes up to 90% of encoding time [5]. Despite
the fact that HEVC provides a simple inter prediction process, the
overhead of the ME block employed is larger compared with
H.264/MPEG-4. Consequently, it is expected to increase the
complexity of HEVC encoder [6]. Typically, as in H.264/MPEG-4
AVC, for every prediction block (PB) in HEVC, the block
matching algorithm (BMA) finds the best matching block within a
certain search window. Full search algorithm is the superior
technique to perform motion compensation in terms of
compression ratio, bit rate (BR) and peak signal-to-noise ratio
(PSNR). Many fast algorithms were proposed to reduce the
computational complexity consumed by the full search.
Nevertheless, most of these algorithms provide a significant
degradation in the bit rate compression or the PSNR or both. In
addition, most of them are not suitable for the hardware
implementation level [7].
Aiming to provide an efficient solution for time, performance and
hardware saving of the bottleneck ME block, we focus on the search
methodology. We first propose a fast centre search algorithm (FCSA)
that modifies the search to start from the centre of the block rather
than the edges. Another algorithm is proposed to speed-up the
search by customising the window size according to the content of
the video sequence. This algorithm is called adaptive search
window algorithm (ASWA). Finally, the centre adaptive search
algorithm (CASA), which is a combination between the first two
algorithms, is proposed. Our proposed algorithms achieve fewer
search points than full search and are comparable with other fast
searches. CASA reduces time complexity by a noteworthy amount,
on average it is 2.1 times faster.
This paper is organised as follows: The principle of operation of
the HEVC ME block is explained in Section 2. In Section 3, the
related work and some fast search algorithms are described. In
Section 4, the details of the proposed algorithms are presented.
Section 5 is dedicated to the experimental results and comparison
with related algorithms. Finally, Section 6 concludes the paper.
2 HEVC ME theory
ME represents paradigm shift in video coding technology. It enables
the encoder to make an efficient use of the temporal redundancies
within a video sequence. The core task of ME unit is to search for
the best matched block in the reference frame within a certain
search window. One of the innovative modifications in HEVC is
to allow variable size coding block to improve the ME quality. For
each CU, the size of its luma CB can be defined as 2N ×2N,
where N ɛ {4, 8, 16, 32}. Code blocks can be further divided into
one, two or four PBs.
IET Image Processing
Research Article
IET Image Process., 2016, Vol. 10, Iss. 6, pp. 438–447
438
&
The Institution of Engineering and Technology 2016
资源评论
码流怪侠
- 粉丝: 2w+
- 资源: 424
最新资源
- GSDML-V2.3-TURCK-BL20-E-GW-EN-20140826-010300.xml
- 基于串行并行ADMM算法的主从配电网分布式优化控制研究 关键词：ADMM 串行并行算法 主动配电网 无功优化 分布式优化 参考文献： 1 参考《主动配电网分布式无功优化控制方法》配电网优化模型
- SAP ERP系统中信用控制配置详解及应用场景分析
- gsdml-v2.25-auto-sm877-pnt-20221215.xml
- 钢筋弯曲折断机step全套技术资料100%好用.zip
- gsdml-v2.31-leuze-bcl348i-20150923.xml
- 钢构复合板打胶机sw16可编辑全套技术资料100%好用.zip
- GSDML-V2.31-Pepperl+Fuchs-PXV100-20171030.xml
- 高压水阀装配生产线sw17可编辑全套技术资料100%好用.zip
- gsdml-v2.31-siemens-sinamics-g120s-vector-20170904.xml
- gsdml-v2.33-sick-dl100hf-20190312.xml
- 环模制粒机step全套技术资料100%好用.zip
- GSDML-V2.34-SICK-Lector63x-1P-20190118.xml
- 基于java的springcloud房产销售平台设计新版源码+数据库+说明
- 基于java的滴答拍摄影项目设计新版源码+数据库+说明
- 基于java的线上教学平台设计新版源码+数据库+说明
资源上传下载、课程学习等过程中有任何疑问或建议，欢迎提出宝贵意见哦~我们会及时处理！
点击此处反馈
