FGPA implementations of motion estimation algorithms using Vivado high level synthesis

Abdul Ghani, Firas (2017) FGPA implementations of motion estimation algorithms using Vivado high level synthesis. [Thesis]

[thumbnail of FirasAbdulGhani_10162831.pdf] PDF
FirasAbdulGhani_10162831.pdf

Download (1MB)

Abstract

Joint collaborative team on video coding (JCT-VC) recently developed a new international video compression standard called High Efficiency Video Coding (HEVC). HEVC has 50% better compression efficiency than previous H.264 video compression standard. HEVC achieves this video compression efficiency by significantly increasing the computational complexity. Motion estimation is the most computationally complex part of video encoders. Integer motion estimation and fractional motion estimation account for 70% of the computational complexity of an HEVC video encoder. High-level synthesis (HLS) tools are started to be successfully used for FPGA implementations of digital signal processing algorithms. They significantly decrease design and verification time. Therefore, in this thesis, we proposed the first FPGA implementation of HEVC full search motion estimation using Vivado HLS. Then, we proposed the first FPGA implementations of two fast search (diamond search and TZ search) algorithms using Vivado HLS. Finally, we proposed the first FPGA implementations of HEVC fractional interpolation and motion estimation using Vivado HLS. We used several HLS optimization directives to increase performance and decrease area of these FPGA implementations.
Item Type: Thesis
Additional Information: Yükseköğretim Kurulu Tez Merkezi Tez No: 478651.
Uncontrolled Keywords: HEVC. -- Fractional interpolation. -- Motion estimation. -- High-Level synthesis. -- HEVC. -- Kesirli Aradeğerleme. -- Hareket Tahmini. -- Yüksek Seviye Sentezleme.
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7800-8360 Electronics
Divisions: Faculty of Engineering and Natural Sciences > Academic programs > Electronics
Faculty of Engineering and Natural Sciences
Depositing User: IC-Cataloging
Date Deposited: 09 May 2018 09:21
Last Modified: 26 Apr 2022 10:22
URI: https://research.sabanciuniv.edu/id/eprint/34738

Actions (login required)

View Item
View Item