The Broadcast Knowledge

Video: Towards Measuring Perceptual Video Quality & Why

Posted on 25th November 2019 by Russell Trafford-Jones

In the ongoing battle to find the minimum bitrate for good looking video, automation is key to achieving this quickly and cheaply. However, metrics like PSNR don’t always give the best answers meaning that eyes are still better the job than silicon.

In this talk from the Demuxed conference, Intel’s Vasavee Vijayaraghavan shows us examples of computer analysis failing to identify lowest bitrate leaving the encoder spending many megabits encoding video so that it looks imperceptibly better. Further more it’s clear that MOS – the Mean Opinion Score – which has a well defined protocol behind it continues to produce the best results, though setting up and co-ordinating takes orders of magnitude more time and money.

Vasavee shows how she’s managed to develop a hybrid workflow which combines metrics and MOS scores to get much of the benefit of computer-generated metrics fed into the manual MOS process. This allows a much more targeted subjective perceptual quality MOS process thereby speeding up the whole process but still getting that human touch where it’s most valuable.

Watch now!
Speaker

Vasavee Vijayaraghavan
Cloud Media Solutions Architect,
Intel

Video: Simplifying the Media Workflow

Posted on 22nd November 2019 by Russell Trafford-Jones

IMF is an exchange format for exchanging media between companies. Wrapping up many different versions of a programme or film into one deliverable, this Interoperable Master Format promises to reduce storage costs, to simplify workflows and, of course, to allow any company to deliver to any other.

Niklas Hammarbäck from the Nordic Entertainment Group explains how they have moved their workflows over to IMF and the benefits that has brought. Niklas lays out the problems he was trying to solve – the main one being the many different delivery formats that must be ingested. These differences create complexity and inefficiencies. The talk examines the requirements that the group developed ahead of transforming their workflows; having a single common format, for example.

This leads in to IMF which Niklas compares to baking a cake. The IMF format contains ingredients and a recipe for creating the deliverable. The ingredients in IMF are the video, audio and metadata files and the recipes are also contained in the delivery. This method allows for a video to be delivered once with several audio files. The traditional alternative would be sending the same video four separate times just with different sound.

Niklas goes in to some detail about the contents of an IMF delivery including the CPL files which are the ‘recipes’ for the media ‘ingredients’ giving examples from https://cpl.fishtank.cloud.

The talk finishes with a summary of the benefits, a check against the requirements and what has been achieved and some questions from the audience.

Watch now!

Speakers

Niklas Hammarbäck
Senior Streaming Specialist,
Nordic Entertainment Group

Video: What to do after per-title encoding

Posted on 21st November 2019 by Russell Trafford-Jones

Per-title encoding is a common method of optimising quality and compression by changing the encoding options on a file-by-file basis. Although some would say the start of per-scene encoding is the death knell for per-title encoding, either is much better than the more traditional plan of applying exactly the same settings to each video.

This talk with Mux’s Nick Chadwick and Ben Dodson looks at what per-title encoding is and how to go about doing it. The initial work involves doing many encodes of the same video and analysing each for quality. This allows you to out which resolutions and bitrates to encode at and how to deliver the best video.

Ben Dodson explains the way they implemented this at Mux using machine learning. This was done by getting computers to ‘watch’ videos and extract metadata. That metadata can then be used to inform the encoding parameters without the computer watching the whole of a new video.

Nick takes some time to explain MUX’s ‘convex hulls’ which give a shape to the content’s performance at different bitrates and helps visualise the optimum encoding parameters the content. Moreover, we see that using this technique, we can explore how to change resolution to create the best encode. This doesn’t always mean reducing the resolution; there are some surprising circumstances when it makes sense to start at high resolutions, even for low bitrates.

The next stage after per-title encoding is to segment the video and encode each segment differently which Nick explores and explains how to deliver different resolutions throughout the stream seamlessly switching between them. Ben takes over and explains how this can be implemented and how to chose the segment boundaries correctly, again, using a machine learning approach to analysis and decision making.

Watch now!
Speakers

	Nick Chadwick Software Engineer, Mux
	Ben Dodson Data Scientist, Mux

Video: Adaptive Bit Rate video delivery (MPEG-DASH)

Posted on 20th November 2019 by Russell Trafford-Jones

MPEG-DASH has been in increasing use for many years and while the implementations and versions continue to improve and add new features, the core of its function remains the same and is the topic of this talk.

For anyone looking for an introduction to multi-bitrate streaming, this talk from Thomas Kernen is a great start as he charts the way streaming has progressed from the initial ‘HTTP progressive download’ to dynamic streaming which adapts to your bandwidth constraints.

Thomas explains the way that players and servers talk and deliver files and summarises the end-to-end distribution ecosystem. He covers the fact that MPEG DASH standardises the container description information, captioning and other aspects. DRM is available through the common encryption scheme.

MPD files, the manifest text files, which are the core of MPEG-DASH are next under the spotlight. Thomas talks us through the difference between Media Presentations, Periods, Representations and Segment Info. We then look at the ability to use the ISO BMFF format or MPEG-2 TS like HLS.

The DASH Industry Forum, DASH-IF, is an organisation which promotes the use of DASH within businesses which means that not only do they do work in spreading the word of what DASH is and how it can be helpful, but they also support interoperability. DASH264 is also the output from the DASH-IF and Thomas describes how this specification of using DASH helps with interoperability.

Buffer bloat is still an issue today which is a phenomenon where for certain types of traffic, the buffers upstream and locally in someone’s network can become perpetually full resulting in increased latency in a stream and, potentially, instability. Thomas looks briefly at this before moving on to HEVC.

At the time of this talk, HEVC was still new and much has happened to it since. This part of the talk gives a good introduction to the reasons that HEVC was brought into being and serves as an interesting comparison for the reasons that VVC, AV1, EVC and other codecs today are needed.

For the latest on DASH, check out the videos in the list of related posts below.

Watch now!
Speaker