메뉴 건너뛰기
.. 내서재 .. 알림
소속 기관/학교 인증
인증하면 논문, 학술자료 등을  무료로 열람할 수 있어요.
한국대학교, 누리자동차, 시립도서관 등 나의 기관을 확인해보세요
(국내 대학 90% 이상 구독 중)
로그인 회원가입 고객센터 ENG
주제분류

추천
검색
질문

논문 기본 정보

자료유형
학술대회자료
저자정보
WEI LIU (Konkuk University) KWON SOONYONG (Konkuk University) SEOJIN JANG (Konkuk University) TEAHYEOUNG SON (Konkuk University) YONG BEOM CHO (Konkuk University)
저널정보
대한전자공학회 대한전자공학회 학술대회 2019년도 대한전자공학회 하계종합학술대회 논문집
발행연도
2019.6
수록면
730 - 740 (11page)

이용수

표지
📌
연구주제
📖
연구배경
🔬
연구방법
🏆
연구결과
AI에게 요청하기
추천
검색
질문

초록· 키워드

오류제보하기
With the rapid increase of high-performance embedded devices equipped with high-end processors, a vast amount of multimedia content is being widely consumed in those small devices. In accordance with this trend, standardization of high-efficiency video coding extensions on screen content coding(HEVC-SCC) is being actively advanced. A promising approach to meeting real-time decoding HEVC-SCC with an embedded device requires an accelerator that uses a multi-core-platform-enhanced graphics processing unit(GPU). In this paper, we present a method to optimize the HEVC-SCC scheme with a multi-core ARM processor and the Nvidia GPU, with enables real-time decoding on embedded platforms. Specifically, we design a multi-threaded and CPU-GPU hybrid parallel computing technique with palette mode, which performs in parallel on the multi-core GPU. In the proposed implementation, the selected architectures are ARM Cortex-A9, ARM Cortex-A15, ARM Cortex-A57, Nvidia Maxwell GPU, and Nvidia Pascal GPU. With our multi-threading technique, the processing speed of the proposed method achieves approximately 34.8 and 34.6 fps on average for video sequences of 1024x720 and 1024x768 pixels, respectively, which is 3.6 times faster than all HEVC decoders. Furthermore, the speed can be accelerated up to five times by the proposed GPU-CPU hybrid parallel computing method. Based on various experimental results, we demonstrate that the proposed method ultimately operates at 44.6 fps for video sequences of both 1027x720 and 1024x768 pixels on embedded platforms

목차

Abstract
1. Introduction
2. Proposed solution
3. Experimental results
4. Conclusions
References

참고문헌 (0)

참고문헌 신청

함께 읽어보면 좋을 논문

논문 유사도에 따라 DBpia 가 추천하는 논문입니다. 함께 보면 좋을 연관 논문을 확인해보세요!

이 논문의 저자 정보

최근 본 자료

전체보기

댓글(0)

0