基於支持向量機(SVM)之多視角高效率視訊編碼器(MV-HEVC)快速演算法

Fast Multiview High Efficiency Video Coding (MV-HEVC) algorithm based on Support Vector Machine (SVM)

余懿修(I-Hsiu Yu)

指導教授 ： 林鼎然 ； 邱奕世

中原大學/電機資訊學院/電子工程研究所/碩士(2019年)

https://doi.org/10.6840/cycu201900457

摘要

高效率視訊編碼 (High Efficiency Video Coding)藉由利用四元樹分割編碼單位CU (coding unit)的特性，以對應幀間預測與幀內預測中各種不同大小的區塊尺寸。然而為了找到最佳的CU劃分而進行非常費時又高複雜度的遞迴式全搜尋，是HEVC在實時傳輸上的一大難題。在本文提出的方法中，利用HEVC編碼器進行編碼時所產生的四元樹分割區塊中的多項參數透過支持向量機 (Support Vector Machines)進行運算，目的是在盡量不影響壓縮表現上提早終止四元樹遞迴運算以節省總運算時間。此方法與HEVC標準相比，在時間節省上有著優異的表現，位元率以及PSNR表現上也相當優秀。實驗結果表明在PSNR表現上依SVM導入點不同約有0.19~0.32 dB掉落，在編碼時間上依SVM導入點不同可以節省41~53%，最高能有63%的時間節省，位元率的表現上與HTM16.2標準近乎一致。

關鍵字

高效率視訊編碼 ； 編碼單元 ； 遞迴提早終止 ； 支持向量機

並列摘要

High Efficiency Video Coding (HEVC) uses the characteristics of quadtree -structured coding unit (CU) block to correspond to different sizes of prediction units in INTER and INTRA prediction modes. Using a very time consuming and highly complex recursive full search to find the best CU partition is a huge problem in real-time transmission. In this thesis, we use the multiple parameters generated by the HEVC encoder, and processed by Support Vector Machines (SVM) to early terminate the quaternary tree recursion operation without affecting the compression performance to save the total computation time. The proposed method has a good performance in terms of time savings, bit rate and Peak Signal to Noise Ratio (PSNR) compared to the HEVC standard. The experimental result show that a slight loss in PSNR by 0.19~0.32dB at different SVM start points. In terms of coding time, depending on the SVM start point, the proposed method saves 41~53% of the time, and the highest can be 63%. The bit rate performance is nearly identical to the HTM 16.2 standard.

並列關鍵字

HEVC (High Efficiency Video Coding) ； CU (coding unit) decision ； Early-termination algorithm ； SVM (Support Vector Machines)

參考文獻

[1] G. J. Sullivan, J. Ohm, W.-J. Han, and T. Wiegand, “Overview of the high efficiency video coding (HEVC) standard,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 22, pp. 1649-1668, 2012.

[2] T. Wiegand, J. R. Ohm, G. J. Sullivan, W.-J. Han, R. Joshi, T. K. Tan, and K. Ugur, “Special section on the joint call for proposals on high efficiency video coding (HEVC) standardization,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 20, pp. 1661- 1666, January 2010.

[3] B. Bross, W.-J. Han, J.-R. Ohm, G. J. Sullivan, and T. Wiegand, “ JCTVC- J1003: High efficiency video coding (HEVC) text specification draft 8,” 2012.

[4] I. K. Kim, J. Min, T. Lee, W.-J. Han, and J. Park, “Block partitioning structure in the HEVC standard,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 22, pp. 1697 - 1706, December 2012.

[5] Y. Zhang, H. Wang, and Z. Li, “Fast coding unit depth decision algorithm for interframe coding in HEVC,” Data Compression Conference (DCC), 2013.