IBM RESEARCH SPOTLIGHT
Computational support for Big Data analytics, information extraction, and visualization to support CIVDDD/NSERC project with IBM Spectrum Computing
Aijun An, York University
IBM Canada Ltd.
SOSCIP, IBM Canada Ltd., OCE, NSERC, Province of Ontario
With millions of videos being uploaded daily to video sharing and social media sites, the need for advanced video streaming technology has become essential in sharing information and telling stories.
The capability to generate and stream quality video data wirelessly from media devices such as mobile phones, tablets and computers, for example, requires sophisticated data science. That’s where Profs. Aijun An and Amir Asif’s team comes in. Their research team, consisting of Vida Movahedi and Mufleh Al-Shatnawi, is building a scalable 3D cloud video transcoder, a technology that will provide high-quality, wireless video transcoding.
The project is part of the BRAIN Alliance, a $10.9 million Ontario Research Fund-Research Excellence grant in big data research. Led by York University with OCAD University, Ryerson University and the University of Toronto, as well as 17 private/public sector partners, the initiative was formerly led by researcher Nick Cercone before his untimely passing.
In honor of Cercone’s significant contribution to the field, Prof. An continued as the project lead among a team of York researchers to develop big data applications using IBM’s Spectrum Symphony™, an IBM software product used for distributed computing and big data analytics.
“The goal of the video cloud transcoding project is to design an error-resilient transcoding framework for mobile 3D video streaming, which transcodes an HD 3D video stream to a mobile, scalable 3D video stream,” explained An.
The project “develops a fast, cloud-based solution that can convert high bitrate video data in HD, 4K or 3D representation to a low bit-rate format that is capable of being adjusted to the receiver’s specifications, while preserving the perceived video quality as much as possible,” she explained.
Prof. An cites support from SOSCIP as crucial in the cloud computing resources provided to post-doctoral fellows.
“At York, I have been fortunate to be able to lead a talented research team. Together we have tackled a variety of research problems in data mining and machine learning, and have established ourselves as a strong research group.”
With a patent application underway, the 3D cloud video transcoder may soon offer a new service to the market-place, providing huge benefits for broadcasting and video streaming companies.