Ioannis Katsavounidis is a member of Video Infrastructure team, where he leads technical efforts to improve video quality and holistic quality across all video products at Meta (formerly Facebook). Before joining Meta, he spent 3.5 years at Netflix, contributing to the development and dissemination of VMAF, Netflix's open-source video quality metric, as well as inventing Dynamic Optimizer, a multi-shot-based video perceptual quality optimization framework that brought significant bit rate savings across the spectrum of streaming applications. VMAF and Dynamic Optimizer awarded Netflix two technical Emmys in 2020. He was an associate professor for 8 years at the Department of Electrical and Computer Engineering at the University of Thessaly in Greece, teaching video compression, signal processing and information theory. He was one of the co-founders of Cidana, a mobile multimedia software company in Shanghai, China. He was software manager for advanced video codecs at InterVideo, makers of WinDVD, the most popular DVD player application, in the early 2000s and also spent 4 years working in experimental High Energy Physics in Italy. He is one of the co-chairs for the Statistical Analysis Methods (SAM) and Non-Reference Measurements (NORM) groups at the Video Quality Expert Group (VQEG). He is actively involved in the Alliance for Open Media (AOM) as co-chair of the Software Implementation Working Group (SWIG). He has over 150 publications, including 50 patents. His research interests are in video coding, video quality, streaming applications, and energy-efficient multimedia processing using hardware and software (HW/SW).
"Energy-efficient video processing - How to achieve it and what is the role of future video compression standards?"
"Social media has multiplied the amount of videos that users upload and share with others on the world wide web. At the same time, the amount of energy required to store, process and share videos with users who wish to view them on their screens has increased. Video compression standards have achieved a significant improvement in video quality, or - if we look at it through the lens of telecommunications - have brought about large reductions in the bit rate required to transmit a video, at the same quality. Traditionally, these improvements in video compression technology have come with equally significant - and lately rather disproportionate - computational complexity/energy impacts. We will present the problem of energy-efficient video processing holistically and suggest some ways to address the increasing demand for video processing through system-level optimization and highlighting first-class citizen energy efficiency in future video compression standards."
Take me back to speakers!