Michael unfolded (.com)

In the paper “An adaptive system for real-time scalable video streaming with end-to-end QoS control”, Beilu Shao et al. introduce and evaluate adaptation enabled by H.264/SVC.

Abstract:

This paper presents a real-time adaptive video streaming system based on the latest standardized video codec H.264/MPEG-4 AVC scalable extension (SVC). The system provides a full MPEG-21 media access framework over heterogeneous networks and terminals with end-to-end QoS control and multimedia adaptation based on SVC. This adaptive streaming system is composed of a server with a real-time SVC encoder, an adaptive network node, and a terminal with appropriate feedback of perceptual quality, network conditions and user preferences for adaptation support. The system facilitates a general content adaptation solution to achieve the end-to-end QoS control.

Link: http://infoscience.epfl.ch/record/148589/files/Adaptation_Camera.pdf

If you are interested in more information on this topic, papers from our research group are available here, here, here and here – just to list a few.

Apple just recently posted a Technical Note which describes how to use hardware (graphics card) accelerated H.264/AVC decoding. This will allow much more efficient decoding of H.264/AVC on OS X platforms and may specifically help to achieve better Adobe Flash performance.

This might also facilitate the development of H.264/SVC decoders for this platform, since the base layer of H.264/SVC is H.264/AVC compatible and therefore the decoding efficiency will benefit from hardware support.

Interesting times ahead

Bob Emmerson provides an interesting writeup on how H.264/SVC enables cost-efficient, high-quality video conferencing.

Here is a short excerpt:

“… Regular telepresence solutions need high quality networks and managed services, and this represents around two-thirds of the total cost. Vidyo claims its solution delivers equivalent or even superior quality over a best-effort network such as the Internet. This allows companies to build videoconferencing capabilities on top of their existing infrastructure. Scalable video coding (SVC) is the technology that provides the flexibility in the video stream to intelligently route only the packets needed by the decoding endpoint in order to create an optimal experience… ”

Read the complete article here.

While I’m inclined to to agree that somewhen in the future the above will be true, there are also those more conservative voices, which mention that there are still many hurdles for H.264/SVC to take – one major is of course “legacy devices”.

A post by Tsahi Levent-Levi from Radvision elaborates on this.

Here is a short excerpt:

“…While H.264/SVC is the future, most of the equipment out there still does H.264/AVC (or even H.263–god forbid). This means that an MCU in today’s world is going to work very hard, harder than usual, with the introduction of H.264/SVC endpoints, to connect these to all the “legacy” endpoints that don’t support it…”

Read the complete article here.

If you want to learn more about H.264/SVC, also refer to my earlier post: “Scalable Video Coding – Executive Summary“

John Bartlett is the Vice President of NetForecast. He has written three very interesting blog posts on H.264/SVC. These are easy to understand executive summaries, which will definitely come in useful when I need to explain this part of my work to non-experts.

Scalable Video Coding Solves Video Conferencing Scaling Issues
In this post, John provides a short introduction on layered video coding (with focus on H.264/SVC of course) in the context of video conferencing.

Scalable Video and Packet Loss
In this post, John explains how Forward Error Correction (FEC) can be used more efficiently with H.264/SVC than with traditional, non scalable, codecs, by only protecting the H.264/SVC base layer.

Scalable Video and Heterogeneous Endpoints
In this post, John points out how H.264/SVC enables to service heterogeneous end devices much more efficiently than non scalable codecs, thanks to its scalability features.

Scalable Video and Scaling of Deployment
In this post, John focuses on how SVC helps to overcome a number of issues when trying to deploy video conferencing to every desktop instead of just a few dedicated conference centers.

For a great technical introduction to H.264/SVC, please see “Overview of the Scalable Video Coding Extension of the H.264/AVC Standard“.

For more adaptation-specific information on H.264/SVC, you can access several of my publications on my Curriculum Vitae & Publications page (scroll down to page 5/6 of my CV and click on the “ITEC/TR” links). Additionally, I’m currently organizing a workshop on “Impact of Scalable Video Coding on Multimedia Provisioning” (SVCVision) where you can discuss the latest advances in H.264/SVC with academic and industry experts. We are still accepting contributions (research papers, demos, position papers, …) as well! Additional publications can be found at our research group.

If you are interested in H.264/SVC, please leave a comment or contact me directly.

Paper submission for the SVCVision workshop is now open! If you have any questions or issues, please contact me.

We are looking forward to your contributions!

http://www.mobimedia.org/ws_SVCVision.html

Our paper with title “Towards User-driven Adaptation of H.264/SVC Streams” was accepted for publication at the Workshop on QoE for Multimedia Content Sharing (QoEMCS 2010).

http://dcti.iscte.pt/events/qoemcs/

Reference:
Jordi Ortiz Murillo, Michael Ransburg, Eduardo Martínez Graciá, Michael Sablatschan, Antonio F. Gómez Skarmeta, Hermann Hellwagner, Towards User-driven Adaptation of H.264/SVC Streams, accepted for publication at Workshop on QoE for Multimedia Content Sharing (QoEMCS 2010), Tampere, Finland, June 9, 2010.

Abstract:
H.264/SVC enables runtime-efficient scalability in the spatial, temporal and fidelity dimension. Existing adaptation mechanisms facilitate this to automatically adapt the H.264/ SVC stream to the current usage environment without any user interaction. This paper argues that the Quality of Experience (QoE) of the end user can be enhanced by enabling him to manually adjust the adaptation if he wishes to do so. An approach which enables this is presented and evaluated. It is shown that by facilitating this approach an increased QoE is provided compared to automatic adaptation approaches. Finally, future work indicates the next steps in order to implement this approach.

Our project on Scalable Video Coding Impact on Networks (SCALNET) will exhibit several demos at the Celtic Event in Valencia next week (from April 12 to April 16). We will show several demos in the context of H.264/SVC.

Further information on the Celtic Event, including a list of all exhibiting projects, can be found here.

A brief overview of the SCALNET project can be found below.

We would be very glad to meet you in Valencia!

Download (PDF, 359KB)

Netflix is a subscription service which allows one to stream series and movies directly to your PC, TV or any other “Netflix ready” device. They also have an application for the iPad in the appstore now. Interestingly enough, one of their selling points for the application is that one can “Resume watching where you left off on your TV or computer“.

This is a concept which is also called “Session Mobility“, i.e., switching sessions between different end devices (and potentially end users). When I started doing research in multimedia communication in 2004, the first project which I worked on was DANAE (Dynamic and distributed adaptation of scalable multimedia content in a context-aware environment). One of the application scenarios of the project was session mobility, as can be seen in the DANAE flyer. This session mobility scenario was not only a lab experiment, but actually implemented in a museum (which was one of the DANAE partners) as a user test.

One slight difference was that we already used the (at this time) current version of the (scalable) H.264/SVC codec in DANAE instead of its predecessor H.264/AVC, which (I assume) is used by Netflix.

A DANAE overview paper can be found here.

It’s good to see concepts from our research making their way into products. I’m thrilled to find more such examples.

Here comes a small update on the SVCVision workshop. During the last weeks, additional experts agreed to join the Program Committee, which now comprises 19 well known experts in the multimedia communication field. Additionally, the submission instructions were detailed and now include, e.g., the maximum number of pages for papers. We also already received encouraging e-mails from both industry and academia which show interest in the workshop.

The paper submission system will be opened very soon. We are looking forward to your contributions. If you have any questions, please feel free to contact me at any time.

Call for Papers
==============
Workshop on Impact of Scalable Video Coding on Multimedia Provisioning (SVCVision)

Collocated with MobiMedia – 6th International Mobile Multimedia Communications Conference
6th-8th September 2010 – Lisbon, Portugal

http://www.mobimedia.org/ws_SVCVision.html

Aims and Scope
===============
Scalable Video Coding (SVC) refers to the possibility of removing certain parts of a video bit stream in order to adapt it to a changing usage environment, e.g., end device capabilities, network condition or user preferences. SVC has been an active standardization and research area for at least 20 years, reaching back to H.262/MPEG-2, which offered scalable profiles. However, these previous attempts suffered from a significant loss in coding efficiency as well as a large increase in decoder complexity (and thus energy consumption), which hindered market adoption. Only the most recent attempt, i.e., the SVC extension of H.264/AVC, focuses on avoiding these disadvantages. Since H.264/SVC standardization started in 2003, it has been at the focus of many multimedia research groups.

Today’s increasing variety of end devices (smart phones, tablet PCs, Netbooks, Laptops, PCs, networked HDTVs, …) and the associated multitude of Internet connectivity options (GPRS/EDGE, UMTS, ADSL, PLC, WiMAX, …) provide particular momentum for SVC, which can be easily and pervasively adapted to these various usage environments. SVC also allows end devices to only decode a sub-set of the SVC bit stream, thus enabling in particular mobile end devices to minimize the necessary (processing) power requirements.

This workshop aims to provide a forum for both academic and industrial participants to exchange and discuss recent advancements and future perspectives of SVC.

SVC topics of interest include, but are not limited to:
==========================================
- Robust streaming, error resilience and error concealment
- Streaming in heterogeneous environments
- Peer-to-Peer (P2P) video distribution
- Internet Protocol television (IPTV)
- Energy-efficient video distribution
- Content adaptation (e.g., scaling, rewriting, transcoding) and summarization
- Complexity optimization and new tools for achieving scalability
- Adaptation decision taking & context information
- Storage & file format
- Conditional access & protection
- Novel applications & implementation experiences

Important Dates
==============
Paper Submission:  23. April 2010
Notification:             28. May 2010
Camera Ready:         25. June 2010

All accepted papers will be published in Springer Lecture Notes of ICST (LNICST) series and included in major article indexing services.

Our paper with the title “Towards an improved SVC-to-AVC Rewriter” was accepted for publication at the Second International Conferences on Advances in Multimedia (MMEDIA 2010).

http://www.iaria.org/conferences2010/MMEDIA10.html

Michael Sablatschan, Michael Ransburg, and Hermann Hellwagner, Towards an improved SVC-to-AVC Rewriter, accepted for publication at Second International Conferences on Advances in Multimedia (MMEDIA 2010), Athens/Glyfada, Greece, June 13-19, 2010.

Abstract:
The Scalable Video Coding (SVC) extension of the H.264/AVC (AVC) video coding standard features spatial, quality and temporal scalability achieved by a layered approach. Backwards compatibility with legacy decoding devices is maintained through an H.264/AVC compliant base layer, which represents the lowest quality of an SVC bit- stream. However, it is often desirable to also provide the higher quality layers to legacy H.264/AVC devices. This is achieved by a process commonly known as “bit-stream rewriting”, which allows for an efficient SVC to AVC conversion by exploiting the similarities of the two codecs. This paper introduces an improved version of the existing JSVM reference software rewriter (JSVM-rewriter). The improvements include a better run-time performance and applicability in streaming scenarios. A detailed evaluation provides performance measurements for the improved rewriter and compares it to the existing JSVM-rewriter. The paper concludes on how the rewriter could be further improved.