Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Video Coding for Large-Scale HTTP Adaptive Streaming Deployments: State of the Art and Challenges Ahead

179 views

Published on

ICIP 2018 Innovation Program: Bringing Industry to ICIP: https://2018.ieeeicip.org/InnovationProgram.asp

Published in: Technology
  • Be the first to comment

  • Be the first to like this

Video Coding for Large-Scale HTTP Adaptive Streaming Deployments: State of the Art and Challenges Ahead

  1. 1. Video Coding for Large-Scale HTTP Adaptive Streaming Deployments: State of the Art and Challenges Ahead O C T O B E R 2 0 1 8
  2. 2. 1 Introduction About the Speaker About Bitmovin Motivation HAS – How it works? Software-based Encoding Cloud Encoding Service Managed On-Premise Encoding Challenges Ahead Multi-Bitrate / Multi-Codec Delay / Quality of Experience VR/360 / AI-based HAS 2 5 4 Agenda What about Standards3 Conclusion6
  3. 3. Associate Professor at the Institute of Information Technology (ITEC), Multimedia Communication Group (MMC), Alpen-Adria- Universität Klagenfurt, Austria; Web: http://itec.aau.at/ Co-founder and CIO | Head of Research and Standardization at Bitmovin; Web: https://bitmovin.com/ Research Interests: immersive multimedia communication, streaming, adaptation, and Quality of experience (QoE) Blog: http://blog.timmerer.com/ Twitter: @timse7 LinkedIn: https://www.linkedin.com/in/christiantimmerer/ SlideShare: https://www.slideshare.net/christian.timmerer About the Speaker 3
  4. 4. Software to Solve Complex Video Problems https://bitmovin.com/ About Bitmovin 4 Encoding Player Analytics Massively distributed video encoding that runs everywhere Deliver High Quality Video everywhere Control and present your data the way your team needs it
  5. 5. Bitmovin Solution 5 End to end software and integrations that help our customers deliver cutting edge solutions with more confidence.
  6. 6. Popular services (global) Netflix (26.58%), HTTP Media Stream (24.40%), YouTube (21.30%), Raw M2TS (8.04%), Amazon Prime (5.73%), Twitch (3.45%); all delivered over-the-top (OTT) Forecast: Visual Networking Index (VNI) 2016-2021 IP video traffic will be 82% of all consumer Internet traffic by 2021 (up from 73% in 2016); will grow threefold from 2016 to 2021 Live Internet video will account for 13% of Internet video traffic by 2021; will grow 15-fold from 2016 to 2021 ______________________________________________ More people now subscribe to Netflix (50.85M) than cable TV (48.61M) in the US (Q1 2017) Motivation 6https://multimediacommunication.blogspot.com/2018/10/almost-58-percent-of-downstream-traffic.html
  7. 7. HTTP Adaptive Streaming – How it works 7 Adaptation logic is within the client, not normatively specified by the standard, subject to research and development
  8. 8. Multi-Bitrate Encoding and Representation Switching 8 Contents on the Web Server Request Movie A (200 Kbps) for t=0 Movie A – 200 Kbps Movie A – 400 Kbps Movie A – 1.2 Mbps Movie A – 2.2 Mbps . . . . . . Request Movie A (400 Kbps) for t=16 Request Movie A (800 Kbps) for t=28 Request Manifest for Movie A Movie K – 200 Kbps Movie K – 500 Kbps Movie K – 1.1 Mbps Movie K – 1.8 Mbps . . . . . . Start quickly Keep requesting Improve quality Loss/congestion detection Revamp quality ... . . . Segments Manifest Request Movie A (400 Kbps) for t=2 Request Movie A (800 Kbps) for t=4
  9. 9. A. Bentaleb, B. Taani, A. C. Begen, C. Timmerer, R. Zimmermann, "A Survey on Bitrate Adaptation Schemes for Streaming Media over HTTP," in IEEE Communications Surveys & Tutorials. https://doi.org/10.1109/COMST.2018.2862938 Bitrate Adaptation Schemes 9 Bitrate Adaptation Schemes Client- based Adaptation Bandwidth- based Buffer- based Mixed adaptation Proprietary solutions MDP-based Server- based Adaptation Network- assisted Adaptation Hybrid Adaptation SDN-based Server and network- assisted
  10. 10. Adobe: HTTP Dynamic Streaming (HDS); switched to DASH Apple: HTTP Live Streaming (HLS); RFC 8216, required for iOS Microsoft: Smooth Streaming; switched to DASH, almost.. Standards 10 Source: http://xkcd.com/927/ MPEG Dynamic Adaptive Streaming over HTTP (DASH) Supported by Netflix, YouTube, Bitmovin, etc. MPEG Common Media Application Format (CMAF) The new kid on the block – support for “fragmented mp4 in HLS” DASH/HLS convergence at segment level – open issues with encryption format
  11. 11. Scalability, speed & control Cloud Encoding Service 11
  12. 12. Cloud Encoding Service Features ○ DRM ○ Live Streaming ○ API Clients ○ Supported Storage ○ Input Formats ○ Output Formats ○ HDR ○ Closed Captions & Subtitles ○ Fully Featured Encoding Service 12
  13. 13. Cloud Encoding Service 13
  14. 14. Massively distributed video encoding that runs anywhere Managed On-Premise Encoding 14
  15. 15. Managed On-Premise Encoding ○ Hybrid Workflows ○ DRM ○ Live Streaming ○ API Clients ○ Supported Storage ○ Input Formats ○ Output Formats ○ Closed Captions & Subtitles ○ Fully Featured Encoding Service Features 15
  16. 16. Which video codecs are you currently using? Video Developer Report 2018 16https://bitmovin.com/bitmovin-2018-video-developer-survey-reveals-shifting-technology-landscape/
  17. 17. Which video codecs are you planning to use in 12 months? Video Developer Report 2018 17https://bitmovin.com/bitmovin-2018-video-developer-survey-reveals-shifting-technology-landscape/
  18. 18. Multi-bitrate encoding Speed, bitrate/resolution, quality [cf. Per-Title Bitrate Ladder Tool] Multi-codec ecosystem AVC, HEVC, VVC, VP9, AV1, AV2 [cf. WQ.L4: Video Coding at Scale] Delay Identified as the biggest problem for video developers in 2018 (55% globally, 74% in LATAM) Challenges Ahead 18 Quality of Experience (QoE) is the degree of “delight or annoyance of the user” of an application or service [cf. QUALINET] VR/360-degree Video Tile-based streaming, MPEG OMAF [cf. “A Framework for Adaptive Delivery of Omnidirectional Video” at HVEI’18] AI-based HAS (end-to-end) Encoding – Streaming – Analytics
  19. 19. ○Multimedia Systems Tradeoff [Based on Klara Nahrstedt at IEEE MIPR’18 Retreat] ○Bitmovin now has 100+ employees, 300+ customers worldwide ○https://bitmovin.com/careers Conclusion 19 Quality Content Time Quality of {Content, Service, Experience, Life, …} Content complexity: traditional AV, AR/VR/360, multi-modality/-sensory End-to-end delay, startup delay, channel switching, synchronization, interaction
  20. 20. THANK YOU QUESTIONS? christian.timmerer@bitmovin.com Christian Timmerer Find Bitmovin on:
  21. 21. References ○ A. Bentaleb, B. Taani, A. C. Begen, C. Timmerer and R. Zimmermann, "A Survey on Bitrate Adaptation Schemes for Streaming Media over HTTP," in IEEE Communications Surveys & Tutorials. https://doi.org/10.1109/COMST.2018.2862938 ○ A. Zabrovskiy, C. Feldmann, C. Timmerer, "Multi-codec DASH dataset," Proc. ACM MMSys'18. https://dx.doi.org/10.1145/3204949.3208140 ○ C. Timmerer, A. C. Begen, "A Framework for Adaptive Delivery of Omnidirectional Video," In Electronic Imaging – Human Vision and Electronic Imaging (HVEI), vol. 2018, no. 16, 2018. http://www.itec.aau.at/bib/files/hvei18-framework-adaptive.pdf ○ M. Graf, C. Timmerer, C. Mueller. "Towards Bandwidth Efficient Adaptive Streaming of Omnidirectional Video over HTTP: Design, Implementation, and Evaluation," Proc. ACM MMSys'17. https://doi.org/10.1145/3083187.3084016 ○ R. Grandl, "Using a Per-Title Bitrate Ladder to Optimize Encoding – Try our new Benchmark Tool," https://bitmovin.com/using-per-title-bitrate-ladder-optimize-encoding-try-new-benchmark-tool/ ○ T. Vernitsky, "Bitmovin 2018 Video Developer Survey," https://bitmovin.com/bitmovin-2018-video- developer-survey-reveals-shifting-technology-landscape/ ○ K. Brunnström, et al., "Qualinet white paper on definitions of quality of experience,” Lausanne, Switzerland, Version 1.2, March 2013. http://www.qualinet.eu

×