本篇我們著重在設計、實作並測量 H.264/SVC 解碼器，將此解碼器移植到多核心行動裝置上，並支援 HTTP 影像串流功能。我們實作的解碼器使用多執行緒，充份發揮多核心的特性，除此之外，影像串流伺服器及用戶端皆支援可動態調整的 HTTP 影像串流。我們為了測量解碼器的效能，在 Android 的平版電腦及智慧型手機上進行實驗，播放 H.264/SVC 影片來獲得數據。依照實驗結果顯示， 在多核心的 Android 行動裝置上進行即時解碼是可行的。舉例來說，當播放 960x544 的影片時，解碼器的畫面更新率（FPS，畫面數／秒）可以達到 20.72 FPS；當播放 480x272 的影片時，畫面更新率可 42.03 FPS。除了對解碼器進行測量外，我們另外做了影像串流的實驗，分別在 WiFi 跟 3G 網路下使用 HTTP 影像串流。使用 HTTP 影像串流時， 我們的解碼器最高可達到 42 FPS，以及短暫的初始延遲（約 2.5 秒）。我們進一步擴充我們的測試平台，支援 MPEG-DASH 標準以及導入 SDL 函式庫。MPEG-DASH 套用 SVC 可帶來下列益處：(1) 可以降低串流伺服器的儲存需求；(2) 跟伺服器要求的檔案可以重複使用達到減少伺服器的負擔；以及(3) 可以更快速地回應調整影片畫質的請求。我們對 MPEG-DASH 以及 SDL 進行實驗。實驗結果顯示資料產出量每秒最高可達到 15 Mbits，播放 360x180 的影片，解碼器可達到至少 50 FPS。最後，我們也將已建置的測試平台開放給研究社群使用。

We design, implement, and evaluate an H.264/SVC decoder and an HTTP video streaming client on multi-core mobile devices. The decoder employs multiple decoder threads to leverage multi-core CPUs, and the streaming server/client support adaptive HTTP video streaming. To evaluate the decoder performance, we conduct experiments using real H.264/SVC videos on a tablet and a smart phone running Android 4.0. Our experimental results demonstrate that real-time H.264/SVC decoding is feasible on multi-core mobile devices. For example, for 960x544 and 480x272 videos, our decoder achieves up to 20.72 and 42.03 Frame-Per-Second (FPS), respectively. We also conduct extensive HTTP video streaming experiments over liveWiFi and 3G cellular networks. Our system achieves high frame rate (up to ~42 FPS), and short initial delay (as small as ~2.5 secs). We extend our testbed to support MPEG DASH (Dynamic Adaptive Streaming over HTTP) standard and SDL (Simple DirectMedia Layer) library. The benefits ofMPEG-DASH with SVC are that (i) the storage space requirement is reduced, (ii) segments can be reused to reduce server overhead, and (iii) switching events are performed faster. Last, we conduct experiments using MPEG-DASH standard and SDL library. The results show that the throughput of MPEG-DASH achieves up to ~15 Mbits/s, and our decoder achieves at least 50 FPS for 360x180 videos. We have made our testbed publicly available to the research communities.

中文摘要 i
Abstract ii
1 Introduction 1
2 Related Work 6
3 Background 8
3.1 H.264/SVC Standard . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.2 MPEG-DASH Standard . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
4 System Architecture 14
5 Implementations 18
5.1 Multi-Core Decoder on Android Devices . . . . . . . . . . . . . . . . . . 18
5.1.1 Limitations of Single-Threaded Decoder . . . . . . . . . . . . . . 18
5.1.2 Software Architecture of Our Multi-Core Decoder . . . . . . . . 18
5.1.3 Parallelism Strategy . . . . . . . . . . . . . . . . . . . . . . . . 19
5.1.4 Porting SDL Library as Renderer to Decoder . . . . . . . . . . . 20
5.2 MPEG-DASH Client with SVC Decoder on Android Devices . . . . . . . 21
5.2.1 Architecture of DASH Client . . . . . . . . . . . . . . . . . . . . 21
5.2.2 Supporting H.264/SVC MPD Format . . . . . . . . . . . . . . . 22
5.2.3 SVC Segment Extractor . . . . . . . . . . . . . . . . . . . . . . 22
5.3 Switching Event Handler . . . . . . . . . . . . . . . . . . . . . . . . . . 23
6 Experiments 25
6.1 Multi-core SVC Decoder . . . . . . . . . . . . . . . . . . . . . . . . . . 25
6.1.1 Videos and Setup . . . . . . . . . . . . . . . . . . . . . . . . . . 25
6.1.2 Evaluation Results of SVC Decoder . . . . . . . . . . . . . . . . 25
6.2 Scalable Video Streaming over HTTP . . . . . . . . . . . . . . . . . . . 28
6.2.1 Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
6.2.2 Evaluation Results of HTTP streaming . . . . . . . . . . . . . . 32
6.3 Evaluation of MPEG-DASH Client . . . . . . . . . . . . . . . . . . . . . 33
6.3.1 Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
6.3.2 Evaluation Results of MPEG-DASH Client . . . . . . . . . . . . 33
6.4 Effective SDL Rendering . . . . . . . . . . . . . . . . . . . . . . . . . . 35
6.4.1 Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
6.4.2 Evaluation Results of SDL Rendering . . . . . . . . . . . . . . . 35
7 Conclusion and FutureWork 38
7.1 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
7.2 Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Bibliography 40

[1] User’s guide, 66321B/D mobile communications dc source. http://cp.literature.agilent.com/litweb/pdf/5964-8184.pdf, 2005.
[2] MPEG-DASH - Part 1: Media presentation description and segment formats. http://www.iso.org/iso/home/store/catalogue_tc/catalogue_detail.htm?csnumber=65274, 2011.
[3] Mplayer-Android project page. https://github.com/ajeet17181/mplayer-android, 2011.
[4] Scalable video coding (SVC video)–Radvision. http://www.radvision.com/Solutions/Video-Communications-Technology/Scalable-Video-Coding/, 2012.
[5] Video conferencing technology platform–Vidyo. http://www.vidyo.com/technology/, 2012.
[6] Video library and tools. http://nsl.cs.sfu.ca/wiki/index.php/Video_Library_and_Tools, 2012.
[7] Opern Source libdash library page. https://github.com/bitmovin/libdash, 2013.
[8] GPAC open source multimedia framework. http://gpac.wp.mines-telecom.fr/, 2014.
[9] Multi-Core SVC Decoder on Android page. https://github.com/nmsl/svc_android_project, 2014.
[10] Cisco visual networking index: Global mobile data traffic forecast update, 2014–2019. http://www.cisco.com/c/en/us/solutions/collateral/service-provider/visual-networking-index-vni/white_paper_c11-520862.html, 2015.
[11] M. Blestel and M. Raulet. Open SVC decoder: A flexible SVC library. In Proc. of ACM Multimedia’10, pages 1463–1466, Firenze, Italy, October 2010.
[12] J. Chong, N. Satish, B. Catanzaro, K. Ravindran, and K. Keutzer. Efficient parallelization of H.264 decoding with macro block level scheduling. In Proc. of IEEE International Conference on Multimedia and Expo (ICME’07), pages 1874–1877, Beijing, China, July 2007.
[13] Y. S. de la Fuente, T. Schierl, C. Hellge, T. Wiegand, D. Hong, D. D. Vleeschauwer, W. V. Leekwijck, and Y. L. Lou´edec. idash: Improved dynamic adaptive streaming over http using scalable video coding. In Proc. of ACM Multimedia Systems (MMSys’11), pages 257–264, San Jose, CA, February 2011.
[14] ITU-T Study Group 9. Subjective video quality assessment methods for multimedia applications. ITU Series P: Audiovisual quality in multimedia services, 1999.
[15] Y. Li, C. Chen, T. Lin, C. Hsu, Y. Wang, and X. Liu. An end-to-end testbed for scalable video streaming to mobile devices over http. In Proc. of IEEE Conference on Multimedia and Expo (ICME’13), San Jose, CA, July 2013.
[16] C. M¨ueller, D. Renzi, S. Lederer, S. Battista, and C. Timmerer. Using scalable video coding for dynamic adaptive streaming over HTTP in mobile environments. In Proc. of European Signal Processing Conference (EUSIPCO’12), pages 2208–2212, Bucharest, Romania, August 2012.
[17] C. M¨uller and C. Timmerer. A test-bed for the dynamic adaptive streaming over HTTP featuring sessionmobility. In Proc. of ACMMultimedia Systems (MMSys’11), pages 271–276, San Jose, CA, February 2011.
[18] H. Richter, B. Stabernack, and E. Muller. Adaptive multithreaded H.264/AVC decoding. In Proc. of Asilomar Conference on Signals, Systems, and Computers (Asilomar’09), pages 886–890, Pacific Grove, CA, November 2009.
[19] H. Schwarz, D. Marpe, and T. Wiegand. Overview of the scalable video coding extension of the H.264/AVC standard. IEEE Transactions on Circuits and Systems for Video Technology, 17(9):1103–1120, September 2007.
[20] C. Sieber, T. Hossfeld, T. Zinner, P. Tran-Gia, and C. Timmerer. Implementation and user-centric comparison of a novel adaptation logic for dash with svc. In Proc. of IEEE Integrated Network Management (IM’13), pages 1318–1323, Ghent, Belgium, May 2013.
[21] W. Song, D. Tjondronegoro, and M. Docherty. Saving bitrate vs. pleasing users: Where is the break-even point in mobile video quality? In Proc. of ACM Multimedia’11, pages 403–412, Scottsdale, AZ, November 2011.
[22] G. Sullivan, G. Bjontegaard, and A. Luthra. Overview of the H.264/AVC video coding standard. IEEE Transactions on Circuits and Systems for Video Technology, 13(7):560–576, July 2003.