Webinar: How to Prove Value with AI and Machine Learning

Today, 2nd June 2020. 17:00 CET | 16:00 BST Register to watch today!

We’ve seen AI entering our lives in many ways over the past few years and we know that this will continue. Artificial Intelligence and Machine Learning are techniques that are so widely applicable they will touch all aspects of our lives before too many more years have passed. So it’s natural for us to look at the broadcast industry and ask “How will AI help us?” We’ve already seen machine learning entering into codecs and video processing showing that up/downscaling can be done better by machine learning than with the traditional ‘static’ algorithms such as bicubic, lanczos and nearest neighbour. This webinar examines the other side of things; how can we use the data available within our supply chains and from our viewers to drive efficiencies and opportunities for better monetisation?

There isn’t a strong consensus on the difference between AI and Machine learning. One is that that Artificial Intelligence is a more broad term of smart computing. Others say that AI has a more real-time feedback mechanism compared to Machine Learning (ML). ML is the process of giving a large set of data to a computer and giving it some basic abilities so that it can learn for itself. A great example of this is the AI network monitoring services available that look at all the traffic flowing through your organisation and learn how people use it. It can then look for unusual activity and alert you. To do this without fixed thresholds (which for network use really wouldn’t work) is really not feasible for humans, but computers are up to that task.

For conversations such as this, it usually doesn’t matter how the computer achieves it, AI, ML or otherwise. The points how can you simplify content production? How can you get better insights into the data you have? How can you speed up manual tasks?

David Short from IET Media moderates this session with Steve Callanan who’s company WIREWAX is working to revolutionise video creation, asset management and interactive video services joined by Hanna Lukashevich from Fraunhofer IDMT (Institute for Digital Media Technology) who uses machine learning to understand and create music and sound. Grant Franklin Totten completes the panel with his experience at Al Jazeera who have been working on using AI in broadcast since 2018 as a way to help maintain editorial and creative compliance as well as detecting fake news and bias checking.

Register to watch today!

David Short Moderator: David Short
Vice Chair,
IET Media Technical Network
Steve Callanan Steve Callanan
Hanna Lukashevich Hanna Lukashevich
Head of Semantic Music Technologies,
Fraunhofer IDMT
Grant Franklin Totten Grant Franklin Totten
Head of Media & Emerging Platforms,
Al Jazeera Media Network

Video: OTT Fundamentals & hands-on video player lab

Whilst there are plenty of videos explaining the basics streaming, few of them talk you through the basics of actually implementing a video player on your website. The principles taught in this hands-on Bitmovin webinar are transferable to many players, but importantly at the end of this talk you’ll have your own implementation of a video player which you can make in real time using their remix project at glitch.com which allows you to edit code and run it immediately in the browser to see your changes.

Ahead of the tutorial, the talk both explains the basics of compression and OTT led by Kieran Farr, Bitmovin’s VP of marketing and Andrea Fassina, Developer Evangelist. Andrea outlines a simplified OTT architecture where he looks at the ‘ingest’ stage which, in this example, is getting the videos from Instagram either via the API or manually. It then looks at the encoding step which compresses the input further and creates a range of different bitrates. Andrea explains that MPEG standards such as H.264, H.265 are commonly used to do this making the point that MPEG standards typically require royalty payments. This year, we are expecting to see VVC released by MPEG (H.266).

Andrea then explains the relationship between resolution, frame rate and file sizes. Clearly smaller files are better as they require less time to download leading to quicker downloads so faster startup times. Andrea discusses how the resolutions match the display resolutions with TVs having 1920×1080 resolution or 2160×3840 resolution. Given that higher resolutions have more picture detail, there is more information to be sent leading to larger file sizes.

Source: Bitmovin https://bit.ly/2VwStwC

When you come to set up your transcoder and player, there are a number of options you need to set. These are determined by these basics, so before launching into the code, Andrea looks further into the fundamental concepts. He next looks at video compression to explain the ways in which compression is achieved and the compromises within. Andrea starts from the first MJPEG codecs where each frame was its own JPEG image and they simply animated from one JPEG to another to show the video – not unlike animated GIFs used on the internet. However by treating each frame on its own ignores a lot of compression opportunity. When looking at one frame to the next, there are a lot of parts of the image which are the same or very similar. This allowed MPEG to step up their efforts and look across a number of frames to spot the similarities. This is typically referred to as temporal compression as is it uses time as part of the process.

In order to achieve this, MPEG splits all frames into blocks, squares in AVC, which are called macro blocks which be compared between frames. They then have 3 types of frame called ‘I’, ‘P’ and ‘B’ frames. The I frames have a complete description of that frame, similar to a JPEG photograph. P frames don’t have a complete description of the frame, rather they some blocks which have new information and some information saying that ‘this block is the same as this block in this other frame. B frames have no complete new image parts, but create the frame purely out of frames from the recent future and recent past; the B stands for ‘bi-directional’.

Ahead of launching into the code, we then look at the different video codecs available. He talks about AVC (discussed in detail here), HEVC (detailed in this talk) and compares the two. One difference is HEVC uses much more flexible macro block sizes. Whilst this increases computational complexity, it reduces the need to send redundant information so is an important part of the achieving the 50% bitrate reduction that HEVC typically shows over AVC. VP9 and AV1 complete the line-up as Andrea gives an overview of which platforms can support these different codecs.

Source: Bitmovin https://bit.ly/2VwStwC

Andrea then introduces the topic of Adaptive bitrate, ABR. This is vital in the effective delivery of video to the home or mobile phones where bandwidth varies over time. It requires creating several different renditions of your content at various bitrates, resolutions and even frame rate. Whilst these multiple encodes put a computational burden on the transcode stage, it’s not acceptable to allow a viewer’s player to go black, so it’s important to keep the low bitrate version. However there is a lot of work which can go into optimising the number and range of bitrates you choose.

Lastly we look at container formats such as MP4 used in both HLS and MPEG-DASH and is based on the file format ISO BMFF. Streaming MP4 is usually called fragmented MP4 (fMP4) as it is split up into chunks. Similarly MPEG2 Transport Streams (TS files) can be used as a wrapper around video and audio codecs. Andrea explains how the TS file is built up and the video, audio and other data such as captions are multiplexed together.

The last half of the video is the hands-on section during which Andrea talks us through how to implement a video player in realtime on the glitch project allowing you to follow along and do the same edits, seeing the results in your browser as you go. He explains how to create a list of source files, get the player working and styled correctly.

Watch now!
Download the presentation

Kieran Farr Kieran Farr
VP of Marketing,
Andrea Fassina Andrea Fassina
Developer Evangelist,

Video: Canon Lenses – A Tale of Three Formats

Lenses are seen by some a black art, by some as a mass of complex physics equations and others who see them as their creative window onto the stories that need to be told. Whilst there is an art behind using lenses, and it’s true making them is complex, understanding how to choose lenses doesn’t require PhD academia.

SMPTE Fellow Larry Thorpe from Canon is here to make the complex accessible as he kicks off talking about lens specifications. He discusses the 2/3-inch image format comparing it with super 35 and full frame. He outlines the specs that are most discussed when purchasing and choosing lenses and shows the balancing act that all lenses are, wanting to maximise sharpness whilst minimising chromatic aberration. On the subject of sharpness, Larry moves on to discussing the way the camera’s ability to sample the video interacts with the lenses ability to capture optical resolution.

Larry considers a normal 1920×1080 HD raster with reference to the physical size of a TV 2/3inch sensor. That works out to be approximately 100 line pairs per millimetre. Packing that into 1mm is tricky if you wish to also maintain quality of the lines. The ability to transfer this resolution is captured by the MTF – the Modulation Transfer Function. This documents the contrast you would see then certain frequencies are viewed through the lens. Larry shows that for a typical lens, this 100 line pairs would have 70% of the original contrast. The higher the frequency, the lower the contrast until it just becomes a flat grey. Larry then looks at a 4K lens showing that it’s needs are 200 line pairs per mm and looking at the MTF, we see that we’re only reaching 50% contrast

Aberrations are important to understand as every lens suffers from them. Larry walks through the 5 classical aberrations, focus and chromatic. To the beginner, chromatic aberrations are, perhaps, the most obvious where colours are seen on the edge of objects, often purple. This is also known as colour fringing. Larry talks about how aperture size can minimise the effect and keeping your image above the 50% contrast limit in the MTF will keep chromatic aberration from being obvious. As a reality check, we then see the limits that have been calculated as limits beyond which it’s simply not possible to improve. Using these graphs we see why 4K lenses offer less opportunity to stop down than HD lenses.

Sharpness zones are zones in lenses optimised for different levels of sharpness. Within the centre, unsurprisingly is the highest sharpness as that’s where most action is. There is then a middle and an outer zone which are progressively less sharp. The reason for this is to recognise that it’s not possible to make the whole image sharp to the same degree. By doing this we are able to create a flatter central zone but with a manage decrease at the corners.

Larry moves on to cover HDR an mentions a recent programme on Fox which was shot in 1080p HDR making the point that HDR is not a ‘4K technology’. He also makes the point that HDR is about the low-lights as well as the specular highlights, so a lens’s ability to be low-noise in the blacks is important an whilst this is not often a problem for SDR, with HDR we are now seeing this coming up more often. For dramas and similar genres, it’s actually very important to be able to shoot whole scenes in low light and Larry shows that the large number of glass elements in lenses is responsible for the low light performance being suboptimal. With up to 50% of light not making it through the lens, this light can be reflected internally and travels around the lens splashing the blacks. Larry explains that coating elements can correct a lot of this and careful choice of the internal surface of the lens mechanisms is also important in minimising such reflections.

Telephoto lenses are lenses which have variable zoom. Larry shows how Canon developed a lens so fully frame a 6 foot athlete from 400 metres away so that they were fully framed on a 2/3″ sensor, but still with a wide angle lens of 60 degrees. With such a long zoom, internal stabilisation is imperative which is done by a very quick active feedback sensor.

So far, Larry has talked about the TV’s standardised 2/3″ image sensor. He now moves on to cover motion format sizes. He shows that for Super 35, you only need 78 line pairs per millimetre which has the knock-on effect of allowing sharper pictures. Next Larry talks about the different versions of ‘full frame’ formats emphasising the creative benefits of larger formats. One is giving a larger field of view which Larry both demonstrates and explains, another is greater sharpness and by having a camera which can choose how much of the sensor you actually use, you can put all sorts of different lenses on. Depth of field is a well known benefit of larger frame formats. The depth of field is much lower which, creatively, is often much desired, though it should be noted that for entertainment shows in TV, that’s much less desirable whilst in films, this is an intrinsic part of the ‘grammar.

As the talk comes to a conclusion, Larry discusses debayering whereby a single sensor has to record red, green and blue. He explains the process and the disadvantages versus separate sensors in larger cameras. As part of this conversion, he shows how oversampling can improve sharpness and avoid aliasing. the talk finishes with an overview of solid storage options

Watch now!

Larry Thorpe Larry Thorpe
National Marketing Executive,
Canon USA Inc.

Video: How to succeed in OTT in 2020

With the streaming market maturing, keeping ahead of competitors requires a layered strategy, particularly for entrants to the market who don’t have Amazon levels of funding. There are success stories out there, so what are the ingredients for success? This recored webinar looks at these questions an shares advice on succeeding in today’s market.

Rahul Patel from Ampere Analysis is first giving an overview of the OTT/streaming market. Within the UK and the US, we now see, he explains, 35% of subscriptions belonging to those 45 and older thus the market is maturing and increasingly offering a wider choice of genres to a wider selection of people. In general, services stand out through their content which is often done through original productions with Netflix now investing almost the same in original programming as the BBC does across all its channels. Rahul shows statistics showing that the percentage of non-US productions is increasing with Netflix and Amazon prime both trying to create programming which is more appealing to their non-US viewers.

An alternative to original programming is to focus your offering. For example, services such as AcornTV focus on UK crime drama, Crunchyroll is focussed on Japanese-made programming and Mubi is 100% films. Though in one market, a very narrow niche may not provide the customers needed to make the service viable, but expanded across all markets, that can all change. Another way services are attracting subscribers, Rahul details, is through bundling either with multiple services being discounted when purchased together, free subscription with hardware purchase or partnerships between big players such as network operators.

As the stacking of services continues to increase, Rahul foresees a future role for aggregators and simplifying the subscription to one payment. Aggregation would involve a single search interface unifying disparate services and could be provided by giants such as Apple and Amazon.

US Industry Avg SCOD Uplifts. Source Reemah Sakaan, BritBox

Next up is Reemah Sakaan from BritBox, which launched in the US in 2017, now in the UK as of 2019 and soon Australia. She explains their journey to market discussing how it can take long time to get rights and forge a true identity, but once that is done, an easy to understand offering can be quickly taken up by viewers. Reemah notes that Covid-19 has pushed a hugh change in viewership both in terms of viewing hours but also subscribers and underlined the importance of their embedded live feed which allows them to go live with special events such as the Royal Wedding along side the traditional VoD offering. This helps them be a unique service and maintain differentiation from competitors.

The last of the presentations is from Simon James from Applicaster. Applicaster’s focus is on providing apps for streaming services. Major issues are in scaling your app on all platforms and trying to manage testing across different OS types and versions. Simon says that the effort needed to keep up with all of this can sap energy and innovation from the team. It also makes it hard to be agile and respond to the market and viewers within the pandemic being a fantastic example of why.

The video ends with a Q&A covering increasing subscriber churn which may lead to annual discounts rather than a monthly subscription. Reemah explains that the first 100 days is key in keeping subscribers. Other questions cover the need for consistent user experience across platforms, approach to expansion and the applicability of your content to your viewers.

Watch now!

Reemah Sakaan Reemah Sakaan
Group Director, SVOD at ITV
Chief Brand and Creative Development Officer, BritBox Global
Rahul Patel Rahul Patel
Ampere Analysis
Simon James Simon James
Head of Sales Engineering,