Showing posts with label WebM. Show all posts
Showing posts with label WebM. Show all posts

Matroska/WebM in MPEG DASH

This document defines how to use Matroska/WebM with the MPEG DASH adaptive streaming system defined in ISO-IEC_23001-6.

Matroska and WebM share the same file structure, Matroska having some extra features. WebM is also restricted to the VP8 video codec and Vorbis audio codec. Only the features present in WebM are taken in consideration in this document.

MPEG-4 AVC/H.264 Video Codecs Comparison

The main goal of this report is the presentation of a comparative evaluation of the quality of new H.264 codecs using objective measures of assessment. The comparison was done using settings provided by the developers of each codec. The main task of the comparison is to analyze different H.264 encoders for the task of transcoding video—e.g., compressing video for personal use. Speed requirements are given for a sufficiently fast PC; fast presets are analogous to real-time encoding for a typical home-use PC.

Overall Conclusions
Overall, the leader in this comparison is x264, followed by DivX H.264, Elecard and MainConcept. The DiscretePhoton encoder demonstrates the worst results among all codecs tested.


Average bitrate ratio for a fixed quality for all categories and all presets (Y-SSIM)


The overall ranking of the codecs tested in this comparison is as follows:

          1. x264
          2. DivX H.264
          3. Elecard
          4. MainConcept
          5. XviD
          6. DiscretePhoton
          • MSE encoder
          • WebM encoder

WebM and Microsoft Expression encoders could not be placed in this list because of their longer encoding time compared with other encoders. The leader in this comparison is x264. Its quality difference (according to the SSIM metric) could be explained by the special encoding option ("tune-SSIM").

The difference between the Elecard and DivX H.264 encoders is almost nothing, and between these encoders and MainConcept is not overly significant, so these encoders tied for second and third in this comparison. This rank is based only on the encoders’ quality results. Encoding speed is not considered here.


x264 settings


Codec Conclusions

• DiscretePhoton - one of the fastest encoder for this comparison, but because of its speed the encoding quality was not very good.

• DivX H.264 - one of comparison leaders, quite balanced encoder with not very big number of parameters, this fact could be comfortable for users. This encoder is designed as a free sample application for DivX Plus HD compliant video encoding, and is a feature-constrained, for-purpose application.

• Elecard - one of comparison leaders, codec with good encoding quality and very flexible settings. Many adjustable encoding settings are provided.

• Microsoft Expression Encoder - encoder with good encoding quality but due to the fact of long initial loading time, the encoding time for Microsoft Expression Encoder is significantly higher than for other encoders.

• MainConcept - good balanced encoder; has many encoding settings that can be adjusted. The results for “Movie” use-case was second, so this codec has a good potential to be one of comparison leaders.

• x264 - one of the best codecs by encoding quality; has very user-friendly predefined presets, as well as many adjustable encoding settings.

• XviD - an MPEG-4 ASP codec; its quality could be very close to or even higher than that of some commercial H.264 standard implementations, especially for encoding “Movie” sequences, but not for “HDTV” sequences.

• WebM - good new non H.264 encoder, it shows good quality but due to the low encoding speed it is not presented in encoders list by quality.

Source: MSU Video Group

H.264 is Still Winning the Codec War

H.264 remains the dominant force in online video, as the video codec now accounts for more than two-thirds of online video, according to a blog post by MeFeedia Thursday. Meanwhile, Google’s WebM format has yet to gain any significant traction after being released a year ago.

H.264's market share continues to widen over competing video formats, as it now accounts for nearly 70 percent of videos indexed by MeFeedia. That’s a huge increase in a very short amount of time, as just last May, when only about 25 percent of videos were available in the H.264 format. And while the percentage growth has slowed in recent quarters, it remains the dominant format for streaming video delivery.



The growth in H.264 encodes is being driven by the adoption of video on tablet devices like the iPad , as well as connected TVs, Blu-ray players and other broadband-enabled video devices. Due to hardware acceleration built into many existing connected device chipsets, H.264 is by far the dominant format for smart TVs and related products.

It’s also an acknowledgement of the strength of the iPad for mobile viewing. There are more than 200 million iOS devices on the market, including 25 million iPads, and H.264 video is the best way to reach those devices. According to MeFeedia, the iPad has the highest engagement among devices, with 40 percent more videos viewed per use than Android, iPhone and desktop users.

While H.264 continues to dominate, the latest numbers on Google’s open-source WebM video format show that it has yet to catch on with publishers. More than a year after its launch, WebM accounts for less than 2 percent of videos indexed, according to MeFeedia. While that is expected to grow — particularly as YouTube continues its process of transcoding all its videos into the WebM format — it’s still a pretty small number for a codec that boasts fairly broad browser adoption and growing support from consumer device manufacturers.

WebM is supported by Firefox, Opera and Google’s Chrome web browsers. In that latter case, in fact, WebM is the default video codec supported for HTML5 video playback, as Google removed support for H.264 in the latest version of Chrome. It’s also gained some hardware backing from consumer electronics manufacturers like Samsung, LG Electronics and Cisco, and has been available on Android devices since the release of Gingerbread.

Despite growing support, it may still take some time before WebM gains the type of hardware acceleration required for broad publisher usage. The good news for Google — and for WebM advocates — is that things can change quickly in the online video market. One need only look at the massive increase in H.264 adoption to see that.

By Ryan Lawler, GigaOM

Initiatives Promise Efficiency Gains For Multi-Screen Service Operations

While the Internet has yet to become an ideal medium for distributing premium content, the good news is there’s much more to come from technology advances that have already gone a long way toward enabling secure, efficient delivery of high-quality entertainment to devices of all descriptions.

In fact, these advances promise not only much better performance for unmanaged over-the- top (OTT) content; they will also streamline efforts of service providers that are developing hybrid managed services where IP-based content is a critical component of multi-device service strategies.

One key development along these lines is progress toward a unifying standard for adaptive streaming under the auspices of the International Organization for Standards’ Moving Picture Experts Group (ISO-MPEG). The MPEG Dynamic Adaptive Streaming over HTTP (DASH) working group has proposed a solution that does away with maintaining separate manifest formats for the two most prominent streaming platforms on the Web – Apple’s HTTP Live Streaming (HLS) and Microsoft’s Smooth Streaming.

At the same time, Hollywood studios through their electronic sell-through initiatives, including most notably the Digital Entertainment Content Ecosystem’s (DECE) UltraViolet platform, have created mechanisms essential to a more interoperable and consumer-friendly online content marketplace. Along with equipping UltraViolet to provide in-the-cloud support for single user accounts, DECE has specified a common file format for all UltraViolet content together with rigorous certification procedures to enable the supply of content protection from multiple vendors.

The Benefits of Adaptive Rate Streaming
When it comes to delivering long-form video over the Internet and bandwidth-constrained access links, one of the great tools invented to compensate for bandwidth fluctuations and other disruptions is Adaptive Rate Streaming (ARS). ARS dynamically adjusts bit rates to changing network and device conditions so that content is delivered at the highest level of quality that can be sustained at any given time without dropping frames or interrupting the flow for buffering.

Thus, for example, for streams targeted to HD sets, ARS bit rates might support an optimum resolution of 1080p when conditions allow and then drop down to rates suited to lower resolutions such as 780p when conditions would cause the stream to stop and buffer if the higher bit rates were maintained. Different ranges of bit rates can be set for different ranges of resolution, depending on the screen sizes of targeted devices.

ARS is a device-driven “stateless” technology, which means the ARS system must match the ARS mode that is supported by the client software running on any given device. At the initiation of a video streaming session by any user on any device, the device is presented live manifest files that define each of the available bit rate profiles for the chosen content. The device signals which profile should be streamed based on what the available access data rate is on the network and how much processing power is available on the device to handle the video stream.

To accommodate fluctuations in network bandwidth and other conditions, the ARS servers rely on feedback from the device to continually update the bit rate profile throughout the session. Every few seconds another file segment or “chunk” is sent to the device at the requested bit rate.

The two most commonly used ARS modes, Apple’s HLS and Microsoft’s Smooth Streaming, have much in common. They both rely on H.264 encoding to set the bit rates. And they employ HTTP (HyperText Transfer Protocol), the streaming protocol that supports the sequence of client request and server response transactions that are the foundation to communications on the World Wide Web. But they are incompatible because they employ different transport container formats.

HLS uses the MPEG Transport Stream format while Smooth Streaming uses the Fragmented MP4 file format for its chunks. The differences in container file sizes impose different time sequences on the client-server communications of the respective ARS systems.

The DASH Initiative
If these differences could be resolved, the fact that H.264 has emerged as the dominant codec by far in IP video communications would make it possible to fashion an ARS standard that would easily fit into the existing video streaming ecosystem. This is precisely what the MPEG DASH working group is attempting to do.

In October 2010, the DASH platform achieved Committee Draft status with expectations that it will reach Final Draft International Standard status by July 2011. The core of the draft is closely aligned with the mobile industry’s 3GPP Adaptive HTTP Streaming (AHS) specification.

AHS, adopted by 3GPP in March 2010, relies on the ISO Base Media File Format ISO/IEC 14496-12, the basis for the MP4 container, to create a uniform approach to streaming video and audio content to mobile devices. DASH has expanded on AHS (now commonly referred to as 3GP-DASH) to create the means by which clients can order ARS segments from servers that use either MP4 encapsulation or MPEG-TS.

This functionality is accomplished through a server-to-client communications mode that delivers a structured collection of data in a format known as the Media Presentation Description to define various segments within the stream, each of which is associated with a uniquely referenced HTTP URL. Altogether these segments describe how a DASH client can use the information to establish a streaming service for the user.

Thus, DASH has been structured to not only provide the means by which the DASH client can order chunks in MP4 or MPEG-TS mode, but to open standardized means of accessing many other functionalities as well. These include support for live streaming as well as progressive download of on-demand content; fast initial startup and seeking; enhanced trick modes and random access capabilities; dual streams for stereoscopic 3D presentations; Multi-view Video Coding used in Blu-ray 3D displays, and dynamic implementation of diverse protection schemes. The platform is designed to work with any HTTP 1.1-compliant server and to support distribution through regular Web infrastructures such as HTTP-based CDNs. This is critical to scaling a common streaming mode to a mass consumer base through use of the embedded base of server farms associated with CDNs worldwide.

Support for DASH
The fact that Microsoft and Apple have joined a global lineup of major firms to bring coherence to adaptive streaming represents a major step forward in the evolution of IP-based distribution of premium content. In a recent blog, Christian Kaiser, vice president of engineering at Netflix, extolled the progress on DASH, saying it has addressed most of the major points required to facilitate a standardized approach to video streaming in conjunction with use of the video playback facility in HTML5, the new multi-media optimized version of the Web’s HyperText Markup Language.

“Since HTML5 includes a facility to embed video playback (the <video> tag), it seems like a natural next step for us to use it for streaming video playback within our HTML5-based user interfaces,” Kaiser said. “However, as of today, there is no accepted standard for advanced streaming through the <video> tag.”

He said such a standard should define acceptable A/V container formats; audio and video codecs; the streaming protocol to be used (such as HTTP); how the streaming protocol adapts to available bandwidth; a way of conveying information about available streams and other parameters to the streaming player module, and a way of exposing these functionalities into HTML5. Netflix has resolved all but the last of these requirements with its own proprietary technology, he noted.

If Netflix could replace its proprietary solution with a standardized ARS platform that met all these conditions, the company and “any other video streaming service could deliver to a standard browser as a pure HTML5 web application, both on computers and in CE devices with embedded browsers,” Kaiser said. “Consumers would benefit by having a growing number of continually evolving choices available on their devices, just like how the web works today for other types of services.”

In Kaiser’s view, DASH provides a solution that would meet all his criteria, with the exception of the last one pertaining to having a way of tying the standard into the HTML5 <video> tag. He said Netflix is engaging with the community to help achieve this integration. Kaiser noted that Netflix intends by early next year to publish a limited subset of the MPEG DASH standard, which will refine the requirements for premium on-demand streaming services. Critically, he noted, the Netflix profile “will take advantage of hooks included in the DASH standard to integrate the DRM technologies that we need to fulfill our contractual obligations to the content providers, thus covering the sixth item [conveying parameters to the player] on our list.”

The fact that DASH creates a way to readily communicate to devices what the specific DRM parameters are for a given piece of content overcomes a major barrier to scaling availability of premium content cost effectively. Today, a content supplier who wants to provide a greater level of security than is supported on any given ARS platform must take specific steps to secure each targeted device with the appropriate DRM client. DASH will allow any DASH-compliant DRM to be implemented automatically with no need for intervention by the content supplier.

Challenges to DASH
Needless to say, DASH will not immediately eliminate all the chaos surrounding competing ARS platform with their variations in codecs, DRMs and streaming profiles. Notably, it remains to be seen how Adobe will adjust to the new standard, although the firm’s move away from sole reliance on proprietary compression with its embrace of H.264 on Flash 9 and 10 and its implementation of an HTTP-based ARS system on Flash 10.1 points to a more open approach in the future. Most recently Adobe indicated it is preparing to provide means by which broadcasters who publish video in Flash can deliver that video to HLS-equipped devices, including the iPhone and iPad.

Another development moving against the tide of ARS standardization is Google’s WebM initiative, which aims to provide what it says is an open-source alternative to H.264 without relying on HTTP streaming. Google says its Chrome browser will abandon support for H.264 in favor of WebM, although its YouTube videos continue to use H.264-based Flash as well as older VP6 compression in videos based on pre-9.0 versions of Flash. Suppliers of the Firefox and Opera browsers, which have never used H.264, will go to WebM in their latest versions.

WebM uses the VP8 video compression format developed as a successor to VP6 by On2, which Google acquired in 2010. It uses the Vorbis codec for audio, which comes out of an open-source project headed by the Xiph.Org Foundation. The transport container used by WebM is based on the Matroska Multimedia Container, the product of another open-source initiative.

Rather than relying on the chunk-based process used by ARS, WebM employs proprietary techniques in conjunction with variable bit rate encoding to effect rate fluctuations within a continuous stream. Describing WebM as an “untried technology,” Ben Schwartz, CTO of Innovation Consulting, in a white paper produced by Harmonic and Verimatrix, questioned the practicality of the variable bit rate (VBR) alternative to ARS. “It is much harder to benefit from caching with this approach, and head-end scalability might become an area of concern,” Schwartz said.

At this nascent stage, it is hard to say how much traction WebM will gain or even whether it will prove to be as royalty-free as Google would like. In February, MPEG-LA, which manages the H.264 patent pool, issued a call for patents related to VP8 in what appeared to be preparations for an assessment as to what royalties might be due from users of VP8.

Microsoft, while not opposed to WebM, indicated it wants to know how Google will provide protection to WebM users against patent claims before it whole-heartedly embraces WebM in its Internet Explorer browser. While Internet Explorer 9 users who install third-party WebM video support on their Windows operating systems will be able to play WebM video in IE9, IE9 will otherwise play HTML5 video in the H.264 format, because H.264 “is a high-quality and widely used video format that serves the Web very well today,” said Dean Hachamovitch, Microsoft’s corporate vice president for IE.

Writing in a blog posted in February, Hachamovitch cited Microsoft’s experiences with previous efforts to offer its Window Media Video (WMV) compression technology (later standardized by the SCTE as VC-1) on an open-source basis as a strong reason to doubt that patent infringement claims won’t be made against VP8. “Asserting openness is not a legal defense,” Hachamovitch said. “The risk question is a legitimate business concern,” he continued. “There are hundreds if not thousands of patents worldwide that read on video formats and codec technologies. Our experience with trying to release WMV for free and open use, and the subsequent claims against Microsoft, support this history as do the cases against JPEG, GIF, and other formats.”

Microsoft offered to work with Google to resolve the risk issues. “Microsoft is willing to commit that we will never assert any patents on VP8 if Google will make a commitment to indemnify us and all other developers and customers who use VP8 in the future,” Hachamovitch said. “We would only ask that we be able to use those patent rights if we are sued first by somebody else. If Google would prefer a patent pool approach, then we would also agree to join a patent pool for VP8 on reasonable licensing terms so long as Google joins the pool and is able to include all other major providers of playback software and devices.”

Clearly, with wide support and the blessing of Apple and Microsoft, DASH is by far the best hope for creating an ARS template that content distributors worldwide can rely on to ease the pain of providing premium content with acceptable quality of experience across all networks and a broader range of device types. As DASH comes into commercial use, distributors employing advanced protection mechanisms will be able to upgrade their distribution networks for content delivered to any DASH compatible client.

UltraViolet
Another major force for simplifying IP-based content distribution is UltraViolet, the electronic sell-through platform developed by the Digital Entertainment Content Ecosystem (DECE). The consortium of more than 60 companies is driven by major studios, including Sony Pictures Entertainment, Fox, Lionsgate, NBC Universal, Paramount, and Warner Bros. Many of the powerhouses in media distribution, retailing and technology are also involved with DECE, including Adobe, Best Buy, BT, CableLabs, Cisco, Comcast, Cox Communications, HP, Huawei, IBM, Intel, LG Electronics, Liberty Global, Microsoft, Motorola, Netflix, Neustar, Nokia, Panasonic, Philips, Samsung Electronics, Thomson and Toshiba.

Once the UltraViolet ecosystem is operational, which is scheduled to happen this summer, consumers will be able to create a cloud-based account with digital rights locker via one of many UltraViolet service providers or through the UltraViolet website. They will then be able to access and manage all of their UltraViolet entertainment for use on all their devices, regardless of where the content was purchased.

The UltraViolet Digital Rights Locker serves as the hub for this new marketplace. The authentication service and account management system allows consumers to access content from multiple registered devices over multiple service outlets operating on broadband and mobile networks. DECE will provide an open API (Application Programming Interface) that allows any Web-enabled storefront, service or device to integrate access to the digital rights locker into its own consumer offering.

To the extent UltraViolet eventually scales to mass usage, the platform will drive uniform approaches to file formatting and DRM that could become a force for standard practices across the entire IP video ecosystem. The UltraViolet Common File Format can be licensed by any participating company to create a consumer offering that will play on any service or device built to DECE specifications. Content providers will be able to encode and encrypt one file type in portable, standard and high definition modes with assurance their files can be accessed by consumers from home storage or the cloud from any registered device.

By mandating a common encryption format and setting rigorous requirements for certifying DRMs for use in UltraViolet, DECE is taking the guesswork out of DRM selection by content distributors and greatly simplifying their ability to accommodate multiple DRMs. Because the file format establishes where in the file sequence various DRM functionalities must be implemented, the cloud-based and local storage systems are able to implement whatever UltraViolet-certified DRM is appropriate to a given piece of content.

DECE has gone a step further toward streamlining content protection by embracing Marlin as one of the approved DRM systems – one of the true multi-vendor DRM systems in the marketplace today, and also security standard at the heart of the Open IPTV Forum’s open standards efforts. Marlin, originated by Panasonic, Philips, Samsung Electronics, Sony and Intertrust Technologies, establishes sophisticated, uniform processes for managing rights on devices in a way that can be integrated into a flexible commercial revenue security offering.

Marlin, which is gaining traction in many parts of the world, was recently selected by the U.K. YouView initiative (previously known as Project Canvas) as the common protection model for all content distributed through the group’s new YouView video store. YouView will aggregate and deliver catchup and on-demand content from the BBC, Channel Five, ITV, Channel 4, BT, Arqiva and TalkTalk to set-top box devices from a range of CE vendors.

Critically, there’s a direct tie-in between Marlin and DASH insofar as DASH has drawn heavily on the work of the Open IPTV Forum. Thus the fact that Marlin is intrinsically compatible with DASH serves as another step forward in the content industry’s efforts to overcome encumbrances to efficient operations resulting from DRM incompatibilities.

By Steve Christian, Screen Plays

Google Hearts Firefox 4 for WebM Video Support

Some folks at Google are getting ready to celebrate as Firefox 4 quickly moves toward 10 million downloads, despite Firefox competing with Google’s Chrome browser. The reason: One of the new features added to Firefox 4 is support for Google’s open video format WebM. The browser release is the biggest boost for WebM since Google open-sourced the format last May, and it could help WebM to finally go mainstream.

Firefox 4 is the first official release to include WebM support. Beta versions of Firefox 4 have been able to play WebM videos for a number of months, but those were limited to a small group of early adopters. Firefox 3.6, on the other hand, was used by around 25 percent of all web users in February, according to the latest data from Statcounter, and all these users are now prompted to update to this week’s release, which will make their machines WebM-capable.


The implications are even bigger when you look at the overall browser market, which has been divided between browsers supporting open video formats and browsers that don’t. Both Safari and IE don’t support WebM and instead use H.264 for HTML5 video playback.

Google’s Chrome added WebM support last year, and Google decided to discontinue support for H.264 entirely earlier this year. Firefox has never supported H.264, but until recently, only supported the inferior Ogg Theora video format for HTML5 video. Opera was the first browser to add support for WebM earlier last year. Combine those two with Firefox 4, and you’re looking at a significant market share, as Mozilla Open Source Evangelist Chris Blizzard pointed out on the WebM blog:

“Firefox accounts for about 30% market share – or nearly a third of all browser users. When you combine that with Chrome and Opera it means that about 50% of internet users will have access to the high-quality WebM codec over the next few months, following the Firefox 4 adoption curve.”

Numbers like these could convince more web publishers to use WebM, or possibly expand some early tests of the format. One candidate for such a move is YouTube, which has been experimenting with WebM as part of its HTML5 trial. This beta test is currently opt-in, and it’s not widely publicized on YouTube’s website.

However, YouTube has been actively converting much of its catalog to WebM, and in November, had already made 80 percent of its regularly requested videos available in the format. With half the browsers accessing the site being capable of WebM playback soon, there’s little reason why YouTube couldn’t promote the format more aggressively.

By Janko Roettgers, GigaOM

Android’s Gingerbread Brings WebM to Mobile Phones

Monday’s release of Android 2.3, code named Gingerbread, is adding support for the WebM open video format to the smart phone platform. Gingerbread users will be able to play WebM videos in their device’s Chrome browser, and Android application developers will be able to make use of WebM as well.

The first version of Android will ship with libvpx 0.9.2, which is a slightly outdated version of WebM. Support for the newest WebM release 0.9.5, code-named Aylesbury, will be pushed out with a maintenance release, according to Google’s WebM product manager John Luther. Users will be able to access every WebM video stream or file with the WebM release included in Gingerbread, but the coming update should help with the format’s playback performance and memory footprint, amongst other things.

Support for Android is an important first step for WebM to gain market share. Google open sourced the video format in May, and it has since been integrated into Firefox, Chrome and Opera. WebM has also gotten more support from video vendors and application developers.

Luther said in November that 80 percent of YouTube’s popular videos are now available in WebM, and Skype’s client started to utilize WebM for its new group video chat functionality. The next step for WebM is to get on devices, and the first chipsets supporting hardware acceleration for WebM are expected to reach the market place in early 2011.

However, it may take a while before many Android users will be able to make use of WebM. Most network operators are notoriously slow to roll out new Android versions. In fact, 56 percent of all Android handsets are still running version 2.1 or older, despite the fact that 2.2 has been available for close to six months now.

By Janko Roettgers, GigaOM