臺灣博碩士論文加值系統

近年來，隨著科技產業的發展，其對高解析度視訊的需求亦在上升。故至至今，仍然值得發展視訊編碼的技術以提升視訊的編碼效能。本文提出兩種方法提升高效率視訊編碼的效能。首先，我們提出一新穎的基於正交匹配追蹤的畫面間殘餘值編碼方法。通過利用正交匹配追蹤去獲取殘餘值得稀疏表達係數從而提升編碼效能。為獲此目的，在畫面內編碼的殘餘值將用於構建一基於紋理複雜度分析的字典。第二種方法，我們在高效率視訊編碼的畫面內編碼中運用卷積神經網路來提升視訊的編碼效能。通過訓練一個基於殘餘值學習機制的卷積神經網路，預測視訊編碼內重建區塊與原始區塊之間的殘餘失真，強化視訊的畫面品質。實驗結果表明本文的方法均能較好的提升視訊編碼效能。

In recent years, an increasing requirement of high resolution video can be observed in the development of technological industry. Nowadays, it is deserved to develop video coding technique unremitting to further improve the coding performance of video. This thesis proposes two methods to improve the coding performance in HEVC. First, we propose a new inter-layer residual coding method based on orthogonal matching pursuit (OMP) to obtain the sparse representation vectors as the transform coefficients. To achieve this purpose, a content adaptive dictionary is constructed in I frame based on the analysis of the coding unit complexity. Then, a novel convolutional neural network (CNN) method is proposed for HEVC intra coding. We train an efficient CNN of a residual learning. In intra frame coding, the proposed CNN predicts the residual for reconstructed blocks to enhance its visual quality. Experimental results show that both methods are achieved favorable coding performance.

Contents
論文審定書 i
誌謝 ii
中文摘要 iii
Abstract iv
List of Figures vi
List of Tables vii
Chapter 1 1
1.1 Overview 1
1.2 Motivation 3
1.3 Contribution 4
1.4 Organization 6
Chapter 2 8
2.1 Coding block structure of HEVC 8
2.2 Intra coding 9
2.3 Residual coding in HEVC 10
2.4 Sparse Coding 11
2.5 Convolutional Neural Network 12
Chapter 3 14
3.1 Coding Unit Complexity (CUC) 14
3.2 Residual Dictionary Construction 15
3.3 Orthogonal Matching Pursuit 16
3.4 Rate-Distortion Optimization 18
3.5 Simulation Results 19
Chapter 4 22
4.2 Residual Learning 24
4.3 Proposed CNN Enhancement Mode for HEVC 25
4.4 Early Termination 26
4.6 Experimental Analysis 31
4.6.1 Training sample extraction 31
4.6.2 Results and Analysis 34
Chapter 5 39
Reference 41

[1]T. Wiegand, G. J. Sullivan, G. Bjontegaard, and A. Luthra, “Overview of the H.264/AVC video coding standard,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 13, no. 7, pp. 560-576, July 2003.
[2]G. J. Sullivan, J.-R. 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, no. 12, pp. 1649-1668, Dec. 2012.
[3]J. A. Tropp, and A. C. Gilbert, “Signal Recovery From Random Measurements Via Orthogonal Matching Pursuit,” IEEE Transactions on Information Theory, vol. 53, no. 12, pp. 4655-4666, Dec. 2007.
[4]Y. Dai, D. Liu, and F. Wu. "A Convolutional Neural Network Approach for Post-Processing in HEVC Intra Coding." International Conference on Multimedia Modeling. Springer, Cham, pp. 28-39, 2017.
[5]W. –S. Park, and M. Kim. "CNN-based in-loop filtering for coding efficiency improvement." Image, Video, and Multidimensional Signal Processing Workshop (IVMSP), pp. 1-5, IEEE, 2016.
[6]S. D. Kim, S. J. Hwang, and M. H. Sunwoo, “Novel residual prediction scheme for hybrid video coding,” in Proceedings of IEEE International Conference on Image Processing, pp. 617-620, 2009.
[7]C. –H. Yeh, C. –W. Lee, S. –J. F. Jiang, Y. -H. Sung, and W. –J. Huang, “Second order residual prediction for HEVC inter coding,” in proceeding of Asia-Pacific Signal and Information Processing Association (APSIPA), Dec. 2014.
[8]J-W. Kang, M. Gabbouj, and C. –C. Kuo, “Sparse/DCT (S/DCT) two-layered representation of prediction residuals for video coding,” IEEE Transactions on Image Processing, Vol. 22, no. 7, pp. 2711-2722, July 2013.
[9]C. Dong, C. C. Loy, K. He, and X, “Tang. Learning a Deep Convolutional Network for Image Super-Resolution,” In Proceedings of European Conference on Computer Vision (ECCV), pp. 184-199, 2014.
[10]C. Dong, Y. Deng, C. C. Loy, and X. Tang, “Compression artifacts reduction by a deep convolutional network,” In Proceedings of the IEEE International Conference on Computer Vision (ICCV). pp. 576-584, 2015.
[11]Y. Li, D. Liu, H. Li, L, Li, F. Wu, H. Zhang, and H. Yang, “Convolutional Neural Network-Based Block Up-sampling for Intra Frame Coding,” IEEE Transactions on Circuits and Systems for Video Technology, vol. pp, no. 99, pp. 1-1, July 2017.
[12]Nair, Vinod, and Geoffrey E. Hinton. "Rectified linear units improve restricted boltzmann machines." Proceedings of the 27th international conference on machine learning (ICML-10). 2010.
[13]K. He, X. Zhang, S. Ren, and J. Sun, "Deep residual learning for image recognition." In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.770-778, 2016.
[14]Y. Lecun, L. Bottou, Y. Bengio and P. Haffner , "Gradient-based learning applied to document recognition." Proceedings of the IEEE, vol. 86, no. 11, pp. 2278-2324, 1998.
[15]Microsoft C++ AMP documentation:
http://msdn.microsoft.com/en-us/library/hh265137.aspx, {2013-03}
[16]HEVC Reference Software HM16.12. (2016 June) [Online]. Available: https://hevc.hhi.fraunhofer.de/trac/hevc/browser/tags/HM-16.12
[17]D.-T. Dang-Nguyen, C. Pasquini, V. Conotter, G. Boato, "RAISE: a raw images dataset for digital image forensics." Proceedings of the 6th ACM Multimedia Systems Conference. ACM, 2015.
[18]Vedaldi, Andrea, and K. Lenc, “MatConvNet: Convolutional neural networks for matlab,” Proceedings of the 23rd ACM international conference on Multimedia. ACM, pp. 689-692, 2015.
[19]G. Bjontegaard, “Calculation of average PSNR differences between RD-curves,” ITU-T, Austin, Texas, United States, technical report VCEG-M33, Apr. 2001.
[20]F. Bossen, “Common HM test conditions and software reference configurations,” JCTVC, Geneva, Switzerland, technical report JCTVC-L1100, Jan. 2013.