Finding innovative approaches to sustainable video delivery

The billions of streaming users around the world get their content from a mere handful of large streaming service providers, each in turn accounting for more bits per second than any other type of internet usage. More than a billion hours of content is consumed on a single streaming platform every single day. Netflix recently claimed that a single hour of streaming a film or drama has the equivalent CO2 emissions as charging 12 smartphones; when you consider that the average American spends nearly 12 hours a day streaming some form of media, that soon mounts up.

The media & entertainment industry is facing an urgent need to reduce its environmental impact, and many of the biggest brands have set ambitious goals of achieving net zero by the end of 2022. In parallel with this mammoth task, the industry must ensure there is no compromising on the impeccable quality and end user experience that today’s audiences demand.

What is needed are truly innovative approaches that are readily available and scalable today, alongside the development of next-generation technologies to ensure that quality and efficiency are continually pushed to new limits.

Looking beyond video compression

The electricity usage for data centres, data transmission and devices, and then on the CO2 emissions associated with each unit of electricity generation, are how we measure and track the carbon footprint of the streaming industry. To reduce the risk of rising energy use and emissions, investments in efficient next-generation computing and communications technologies are needed, alongside continued efforts to decarbonise the electricity supply.

iSIZE believes we need innovative approaches for a brave new world that look beyond the use of standard video compression algorithms. AI-based pre-processing prior to encoding makes the ingested content easier, and more efficient, to encode – this is where we see the biggest gains being made.

The aim of new approaches to video delivery, such as iSIZE’s, is to bring together psychovisual approaches and artificial intelligence to remove video information that is known to be imperceptible by viewers while reaching outstanding compression levels by standard MPEG or AOMedia encoders.

Not all pixels are created equal. With an understanding of how people perceive video content, it becomes possible to remove unnecessary detail in the input video content that incurs significant bitrate overhead in typical video encoders.

The end user experience is a major point of differentiation, so it is vital that the above process creates zero impairment to the visual quality – whether we are sharing videos for family contact or for business, we do not want them to look blurry/noisy or cartoon-like. This is where quantifying visual distortion becomes critical, and it is here that AI and deep psychovisual pre-processing become a gamechanger.

Leveraging neural networks for this type of video pre-processing is ideally suited to GPU operations; computations can be run in a massively parallel architecture. On a small scale it could run on a typical PC or mobile phone, since all such devices now have increased GPU or NPU (neural processing unit) capabilities. Broadcasters and content providers could also use cloud processing to deliver at scale and at resolutions up to Ultra HD. Any processing overhead is more than counterbalanced by the reduction in encoding complexity as well as a significant decrease in the bandwidth requirements for the compressed video.

Removing complexity with AI

An AI engine learns to distinguish perceptually unnoticeable details in the content in an autonomous manner and without requiring any input from the encoder. The result is an increase in the compression efficiency of AVC, HEVC, VP9 and AV1 between 12% to 50% (depending on the use case), as validated by extensive commercial tests and human mean opinion scores obtained with standard protocols like ITU-T P.910 tests.

The other key factor in successfully implementing such technology is to ensure 100% standard compliance and cross-standard/cross-codec applicability; this removes reliance on any standard or format and can be applied to any application, platform, or workflow that must move video data quickly and efficiently. The result is seamless integration without breaking any video coding or streaming standards.

Avoiding additional compute complexity in already complicated workflows is a key issue for many customers. Solutions that increase the efficiency and performance of all the latest codec standards including AVC/H.264, HEVC/H.265, and VP9 typically add significant compute complexity at the same time. iSIZE’s codec-independence and fast execution means the capability to reduce video delivery system bitrate requirements without adding significant complexity. Our approach eliminates the need to wait for new codec standards to be developed and widely adopted – a lengthy process in an industry that is moving at pace.

At iSIZE our belief is that sustainability and greater efficiency for seamless end-user experiences are not mutually exclusive. We are also proponents of implementing standards-based solutions, that are available today, and can begin to reduce the environmental impact of video streaming immediately.