Mass center direction-based decision method for intraprediction in HEVC standard
- PDF / 2,898,683 Bytes
- 16 Pages / 595.276 x 790.866 pts Page_size
- 79 Downloads / 171 Views
ORIGINAL RESEARCH PAPER
Mass center direction-based decision method for intraprediction in HEVC standard Narjes Najafabadi1 · Mohammadreza Ramezanpour2 Received: 3 November 2018 / Accepted: 1 March 2019 © Springer-Verlag GmbH Germany, part of Springer Nature 2019
Abstract Increasing the prediction modes and recursive quad-tree structure, can be the reasons for high compression efficiency in high-efficiency video coding (HEVC) standard. For the same reasons, the computational complexity of HEVC is increased compared with the previous standards. One of the parts that impose a high computational load is the intraprediction unit. In this paper, an improved intraprediction mode decision method is presented to accelerate intracoding of HEVC. A mass center direction-based approach is used to find correlation direction of pixels which can effectively eliminate the prediction modes that have low chance to be selected as the best intramode from the rate-distortion optimization computations. According to the founded correlation direction for 4 × 4 blocks, the depth range of coding units can also be reduced to accelerate the intracoding further. The performance of the proposed method evaluated with different test sequences proposed by the joint collaborative team on video coding. The coding results indicate that the proposed method reduces the encoding time significantly as compared with the HM-16.19 reference software with negligible loss of peak signal-to-noise ratio quality. Keywords Mass center direction · Intracoding · Prediction mode · HEVC
1 Introduction High-resolution television is becoming increasingly common because of advances in capturing and display technologies. Existing video coding standards such as MPEG2, MPEG-4, and H.264 [1] cannot satisfy the demands of high-definition video applications. The Joint Collaborative Team on Video Coding (JCT-VC) co-established by ISO/ IEC and ITU-T developed the next generation of video coding standards; that is, high-efficiency video coding (HEVC) [2]. HEVC uses variable sizes of coding units (CUs), prediction units (PUs), and transform units (TUs). It is a significant improvement regarding many aspects, because it uses a hierarchical quad-tree structure for motion compensation, multi-direction prediction, adaptive pixel compensation, * Mohammadreza Ramezanpour [email protected]; [email protected] Narjes Najafabadi [email protected] 1
Faculty of Computer Engineering, Najafabad Branch, Islamic Azad University, Najafabad, Iran
Department of Computer Engineering, Mobarakeh Branch, Islamic Azad University, Mobarakeh, Isfahan, Iran
2
and a loop filter. Compared with H.264/AVC, it improves the coding efficiency by 100% [3, 4]. HEVC uses a quadtree-structured CU, which allows for a recursive split into four equally sized blocks. In the largest CU (LCU), each CU with different depths is recursively looped through to decide the best CU size and prediction mode according to the rate-distortion optimization (RDO). However, these new technologies are ti
Data Loading...
