Video:Measuring Video Quality with VMAF – Why You Should Care

VMAF, from Netflix, has become a popular tool for evaluating video quality since its launch as an Open Source project in 2017. Coming out of research from the University of Southern California and The University of Texas at Austin, it’s seen as one of the leading ways to automate video assessment.

Netflix’s Christos Bampis gives us a brief overview of VMAF’s origins and its aims. VMAF came about because other metrics such as MS-SSIM and, in particular, PSNR aren’t close enough indicators of quality. Indeed, Christos shows that when it comes to animated content (i.e. anime and cartoons) subjective scores can be very high, but if we look at the PSNR score it can be the same as the PSNR of score another live-action video clip which humans rate a lot lower, subjectively. Moreover, in less extreme examples, Christos explains. PSNR is often 5% or so away from the actual subjective score in either direction.

To a simple approximation, VMAF is a method of bringing out the spatial and temporal information from a video frame in a way which emphasises the types of things humans are attuned to such as contrast masking. Christos shows an example of a picture where artefacts in the trees are much harder to see than similar artefacts on a colour gradient such as a sky or still water. These extraction methods take account of situations like this and are then fed into a trained model which matches the results of the model with the numbers that humans would have given it. The idea being that when trained on many examples, it can correctly predict a human’s score given a set of data extracted from a picture. Christos shows examples of how well VMAF out-performs PSNR in gauging video quality.

 

Challenges are in focus in the second half of the talk. What are the things which still need working on to improve VMAF? Christos zooms in on two: design dimensionality and noise. By design dimensionality, he means how can VMAF be extended to be more general, delivering a number which has a consistent meaning in different scenarios? As the VMAF model has been trained on AVC, how can we deal with different artefacts which are seen with different codecs? Do we need a new model for HDR content instead of SDR and how should viewing conditions, whether ambient light or resolution and size of the display device, be brought into the metric? The second challenge Christos highlights is noise as he reveals VMAF tends to give lower scores than it should to noisy sources. Codecs like AV1 have film-grain synthesis tools and these need to be evaluated, so behaving correctly in the presence of video noise is important.

The talk finishes with Christos outlining that VMAF’s applicability to the industry is only increasing with new codecs coming out such as LCEVC, VCC, AV1 and more – such diversity in the codec ecosystem wasn’t an obvious prediction in 2014 when the initial research work was started. Christos underlines the fact that VMAF is a continually evolving metric which is Open Source and open to contributions. The Q&A covers failure cases, super-resolution and how to interpret close-call results which are only 1% different.

Watch now!
Download the presentation
Speaker

Christos Bampis Christos Bampis
Senior Software Engineer,
Netflix

Video: AV1 Commercial Readiness Panel

With two years of development and deployments under its belt, AV1 is still emerging on to the codec scene. That’s not to say that it’s no in use billions of times a year, but compared to the incumbents, there’s still some distance to go. Known as very slow to encode and computationally impractical, today’s panel is here to say that’s old news and AV1 is now a real-time codec.

Brought together by Jill Boyce with Intel, we hear from Amazon, Facebook, Googles, Amazon, Twitch, Netflix and Tencent in this panel. Intel and Netflix have been collaborating on the SVT-AV1 encoder and decoder framework for two years. The SVT-AV1 encoder’s goal was to be a high-performance and scalable encoder and decoder, using parallelisation to achieve this aim.

Yueshi Shen from Amazon and Twitch is first to present, explaining that for them, AV1 is a key technology in the 5G area. They have put together a 1440p, 120fps games demo which has been enabled by AV1. They feel that this resolution and framerate will be a critical feature for Twitch in the next two years as computer games increasingly extend beyond typical broadcast boundaries. Another key feature is achieving an end-to-end latency of 1.5 seconds which, he says, will partly be achieved using AV1. His company has been working with SOC vendors to accelerate the adoption of AV1 decoders as their proliferation is key to a successful transition to AV1 across the board. Simultaneously, AWS has been adding AV1 capability to MediaConvert and is planning to continue AV1 integration in other turnkey content solutions.

David Ronca from Facebook says that AV1 gives them the opportunity to reduce video egress bandwidth whilst also helping increase quality. For them, SVT-AV1 has brought using AV1 into the practical domain and they are able to run AV1 payloads in production as well as launch a large-scale decoder test across a large set of mobile devices.

Matt Frost represent’s Google Chrome and Android’s point of view on AV1. Early adopters, having been streaming partly using AV1 since 2018 in resolution small and large, they have recently added support in Duo, their Android video-conferencing application. As with all such services, the pandemic has shown how important they can be and how important it is that they can scale. Their move to AV1 streaming has had favourable results which is the start of the return on their investment in the technology.

Google’s involvement with the Alliance for Open Media (AOM), along with the other founding companies, was born out of a belief that in order to achieve the scales needed for video applications, the only sensible future was with cheap-to-deploy codecs, so it made a lot of sense to invest time in the royalty-free AV1.

Andrey Norkin from Netflix explains that they believe AV1 will bring a better experience to their members. Netflix has been using AV1 in streaming since February 2020 on android devices using a software decoder. This has allowed them to get better quality at lower bitrates than VP9 Testing AV1 on other platforms. Intent on only using 10-bit encodes across all devices, Andrey explains that this mode gives the best efficiency. As well as being founding members of AoM, Netflix has also developed AVIF which is an image format based on AV1. According to Andrey, they see better performance than most other formats out there. As AVIF works better with text on pictures than other formats, Netflix are intending to use it in their UI.

Tencent’s Shan Liu explains that they are part of the AoM because video compression is key for most Tencent businesses in their vast empire. Tencent cloud has already launched an AV1 transcoding service and support AV1 in VoD.

The panel discusses low-latency use of AV1, with Dave Ronca explaining that, with the performance improvements of the encoder and decoders along-side the ability to tune the decode speed of AV1 by turning on and off certain tools, real-time AV1 are now possible. Amazon is paying attention to low-end, sub $300 handsets, according to Yueshi, as they believe this will be where the most 5G growth will occur so site recent tests showing decoding AV1 in only 3.5 cores on a mobile SOC as encouraging as it’s standard to have 8 or more. They have now moved to researching battery life.

The panel finishes with a Q&A touching on encoding speed, the VVC and LCEVC codecs, the Sisvel AV1 patent pool, the next ramp-up in deployments and the roadmap for SVT-AV1.

Watch now!
Please note: After free registration, this video is located towards the bottom of the page
Speakers

Yueshi Shen Yueshi Shen
Principle Engineer
AWS & Twitch
David Ronca David Ronca
Video Infrastructure Team,
Facebook
Matt Frost Matt Frost
Product Manager, Chome Media Technologies,
Google
Andrey Norkin Andrey Norkin
Emerging Technologies Team
Netflix
Shan Liu Dr Shan Liu
Chief Scientist & General Manager,
Tencent Media Lab
Jill Boyce Jill Boyce
Intel

Video: CMAF And The Future Of OTT

Why is CMAF still ‘the future’ of OTT? Published in 2018, CMAF’s been around for a while now so what are the challenges and hurdles holding up implementation? Are there reasons not to use it at all? CMAF is a way of encoding and packaging media which then can be sent using MPEG DASH and HLS, the latter being the path Disney+ has chosen, for instance.

This panel from Streaming Media West Connect, moderated by Jan Ozer, discusses CMAF use within Akami, Netflix, Disney+, and Hulu. Peter Chave from Akamai starts off making the point that CMAF is important to CDNs because if companies are able to use just one CMAF file as the source for different delivery formats, this reduces storage costs for consumers and makes each individual file more popular thus increasing the chance of having a file available in the CDN (particularly at the edge) and reducing cache misses. They’ve had to do some work to ensure that CMAF is carried throughout the CDN efficiently and ensuring the manifests are correctly checked.

Disney+, explains Bill Zurat, is 100% HLS CMAF. Benefiting from the long experience of the Disney Streaming Services teams (formerly BAMTECH), but also from setting up a new service, Disney were able to bring in CMAF from the start. There are issues ensuring end-device support, but as part of the launch, a number were sunsetted which didn’t have the requirements necessary to support either the protocol or the DRM needed.

Hulu is an aggregator so they have strong motivation to normalise inputs, we hear from Hulu’s Nick Brookins. But they also originate programming along with live streaming so CMAF has an important to play on the way in and the way out. Hulu dynamically regenerates their manifests so can iterate as they roll out easily. They are currently part the way through the rollout and will achieve full CMAF compatibility within the next 18 months.

The conversation turns to DRM. CMAF supports two methods of DRM known as CTR (adopted by Apple) and CBC (also known as CBCS) which has been adopted by others. AV1 supports both, but the recommendation has been to use CBC which appears have been universally followed to date explains Netflix’s Cyril Concolato. Netflix have been using AV1 since it was finalised and are aiming to have most titles transitioned by 2021 to CMAF.

Peter comments from Akamai’s position that they see a number of customers who, like Disney+ and Peacock, have been able to enter the market recently and move straight into CMAF, but there is a whole continuum of companies who are restricted by their workflows and viewer’s devices in moving to CMAF.

Low latency streaming is one topic which invigorates minds and debates for many in the industry. Netflix, being purely video on demand, they are not interested in low-latency streaming. However, Hulu is as is Disney Streaming Services, but Bill cautions us on rushing to the bottom in terms of latency. Quality of experience is improved with extra latency both in terms of reduced rebuffering and, in some cases, picture quality. Much of Disney Streaming Services’ output needs to match cable, rather than meeting over-the-air latencies or less.

The panel session finishes with a quick-fire round of questions from Jan and the audience covering codec strategy, whether their workflows have changed to incorporate CMAF, just-in-time vs static packaging, and what customers get out of CMAF.

Watch now!
Speakers

Cyril Concolato Cyril Concolato
Senior Software Engineer,
Netflix
Peter Chave Peter Chave
Principal Architect,
Akamai
Nick Brookins Nick Brookins
VP, Platform Services Group,
Hulu
Bill Zurat Bill Zurat
VP, Core Technology
Disney Streaming Services
Jan Ozer Moderator: Jan Ozer
Contributing Editor, Streaming Media
Owner, StreamingLearningCenter.com

Video: AV1 – A Reality Check

Released in 2018, AV1 had been a little over two years in the making at the Alliance of Open Media founded by industry giants including Google, Amazon, Mozilla, Netflix. Since then work has continued to optimise the toolset to bring both encoding and decoding down to real-world levels.

This talk brings together AOM members Mozilla, Netflix, Vimeo and Bitmovin to discus where AV1’s up to and to answer questions from the audience. After some introductions, the conversation turns to 8K. The Olympics are the broadcast industry’s main driver for 8K at the moment, though it’s clear that Japan and other territories aim to follow through with further deployments and uses.

“AV1 is the 8K codec of choice” 

Paul MacDougall, Bitmovin
 CES 2020 saw a number of announcements like this from Samsung regarding AV1-enabled 8K TVs. In this talk from Google, Matt Frost from Google Chrome Media explains how YouTube has found that viewer retention is higher with VP9-delivered videos which he attributes to VP9’s improved compression over AVC which leads to quicker start times, less buffering and, often, a higher resolution being delivered to the user. AV1 is seen as providing these same benefits over AVC without the patent problems that come with HEVC.

 
It’s not all about resolution, however, points out Paul MacDougall from BitMovin. Resolution can be useful, for instance in animations. For animated content, resolution is worth having because it accentuates the lines which add intelligibility to the picture. For some content, with many similar textures, grass, for instance, then quality through bitrate may be more useful than adding resolution. Vittorio Giovara from Vimeo agrees, pointing out that viewer experience is a combination of many factors. Though it’s trivial to say that a high-resolution screen of unintended black makes for a bad experience, it is a great reminder of things that matter. Less obviously, Vittorio highlights the three pillars of spatial, temporal and spectral quality. Temporal refers to upping the bitrate, spatial is, indeed, the resolution and spectral refers to bit-depth and colour-depth know as HDR and Wide Colour Gamut (WCG).

Nathan Egge from Mozilla acknowledges that in their 2018 code release at NAB, the unoptimized encoder which was claimed by some to be 3000 times slower than HEVC, was ’embarrassing’, but this is the price of developing in the open. The panel discusses the fact that the idea of developing compression is to try out approaches until you find a combination that work well. While you are doing that, it would be a false economy to be constantly optimising. Moreover, Netflix’s Anush Moorthy points out, it’s a different set of skills and, therefore, a different set of people who optimise the algorithms.

Questions fielded by the panel cover whether there are any attempts to put AV1 encoding or decoding into GPU. Power consumption and whether TVs will have hardware or software AV1 decoding. Current in-production AV1 uses and AVC vs VVC (compression benefit Vs. royalty payments).

Watch now!
Speakers

Vittorio Giovara Vittorio Giovara
Manager, Engineering – Video Technology
Vimeo
Nathan Egge Nathan Egge
Video Codec Engineer,
Mozilla
Paul MacDougall Paul MacDougall
Principal Sales Engineer,
Bitmovin
Anush Moorthy Anush Moorthy
Manager, Video and Image Encoding
Netflix
Tim Siglin Tim Siglin
Founding Executive Director
Help Me Stream, USA