High definition content is becoming an integral part of global broadcasting, setting the standard for streaming, satellite, cable and terrestrial transmissions. More importantly, it is generating extremely large amounts of data that need to be delivered to a large number of users. Up to now, the distribution of conventional media has only been made possible by the application of compression systems that reduce the size of content sufficiently to be stored and distributed on a mass scale. Continuing this trend and exploiting recent breakthroughs in compression technology, the emerging High Efficiency Video Coding (HEVC) standard provides a base for handling very high resolution content. Underpinning the capabilities of HEVC are computationally expensive, extremely complex and resource demanding algorithms. Consequently, in its current state, HEVC is unfeasible for practical applications involving very high definition content. The main objective of this project is to research and develop economically viable, i.e., resource constrained, technology for very high definition content handling. The goal is to facilitate storage and transmission of the huge amounts of data contained in very high resolution content, paving a way for the provision of much greater quality of experience, facilitating quick deployment in a variety of important applications and sectors including broadcasting, digital cinema, medical, biological and defence.

HEVC is a new video compression standard due for completion by JCT-VC in January 2013. It is the first significant development in video compression since H.264 was created a decade ago. Thanks to more sophisticated video processing and compression algorithms, HEVC is capable of delivering video of equivalent quality as ten-year-old predecessor H.264 while using only half the bitrate. HEVC is many times more complex than H.264 making implementation correspondingly much more difficult. Complexity is especially felt on the encoder side where an encoder has to decide between many more permutations of spatial subdivisions and multiple encoding modes. HEVC encoding remains a challenge to the industry. Offline, non-real-time encoding is possible albeit extremely slow. For some applications, notably broadcast and video conferencing, real-time and low-latency encoding is necessary and “brute-force”, supercomputing approaches are being used here. Parabola believes that it is possible to use a mix of innovative optimisation and algorithm approaches to implement HEVC encoding on commodity hardware whilst retaining or even beating HEVC's “50% bitrate” headline advantage. This aim of this project is to prove that it is possible to achieve an eventual goal of real-time, high definition encoding on contemporary consumer hardware without sacrificing the advantages of HEVC. The project will develop, evaluate, refine and demonstrate shortcut and heuristic algorithms that, when combined with conventional software optimization, realise the promise of HEVC on modest computers and devices. The key objectives of this project are: • to demonstrate feasibility of real-time software encoding on consumer hardware • to generate valuable intellectual property in the field of video compression • to augment Parabola’s existing HEVC technology to support fast encoding