An efficient CU partition algorithm for HEVC based on improved Sobel operator

Xuebin Sun; Xiaodong Chen; Yong Xu; Gang Sun; Yunsheng Yang

doi:10.1117/12.2302920

10 April 2018 An efficient CU partition algorithm for HEVC based on improved Sobel operator

Xuebin Sun, Xiaodong Chen, Yong Xu, Gang Sun, Yunsheng Yang

Proceedings Volume 10615, Ninth International Conference on Graphic and Image Processing (ICGIP 2017); 1061544 (2018) https://doi.org/10.1117/12.2302920
Event: Ninth International Conference on Graphic and Image Processing, 2017, Qingdao, China

Abstract

As the latest video coding standard, High Efficiency Video Coding (HEVC) achieves over 50% bit rate reduction with similar video quality compared with previous standards H.264/AVC. However, the higher compression efficiency is attained at the cost of significantly increasing computational load. In order to reduce the complexity, this paper proposes a fast coding unit (CU) partition technique to speed up the process. To detect the edge features of each CU, a more accurate improved Sobel filtering is developed and performed By analyzing the textural features of CU, an early CU splitting termination is proposed to decide whether a CU should be decomposed into four lower-dimensions CUs or not. Compared with the reference software HM16.7, experimental results indicate the proposed algorithm can lessen the encoding time up to 44.09% on average, with a negligible bit rate increase of 0.24%, and quality losses lower 0.03 dB, respectively. In addition, the proposed algorithm gets a better trade-off between complexity and rate-distortion among the other proposed works.

Citation Download Citation

Xuebin Sun, Xiaodong Chen, Yong Xu, Gang Sun, and Yunsheng Yang "An efficient CU partition algorithm for HEVC based on improved Sobel operator", Proc. SPIE 10615, Ninth International Conference on Graphic and Image Processing (ICGIP 2017), 1061544 (10 April 2018); https://doi.org/10.1117/12.2302920

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available