Mariana Afonso
Netflix (USA)
The journey towards better video compression for VOD streaming
Language: English
Date: October 28th / 15h00 – 16h15 (UTC-3)
Abstract
Netflix is one of the leaders in video on demand (VOD) streaming, providing entertainment to millions of members worldwide from an extensive catalog of shows and movies. Providing the best quality of experience through a variety of devices and network conditions requires optimized encoding algorithms and strategies, reliable quality estimations and improved codecs. In this workshop, we will provide an overview of the world of video streaming at Netflix, from the high-quality footage provided by the studio all the way to the devices streaming the content at home. We will also present interesting challenges and showcase recent internal developments in the fields of video compression, encoding and video quality assessment. In addition, efforts towards the adoption of newer codecs and research towards improved royalty-free video codecs within the Alliance for Open Media (AOM) will be presented.
Biography
Mariana Afonso received her B.Sc./M.Sc. in Electrical and Computer Engineering from the University of Porto, Portugal, in 2015. She later received her Ph.D. in Electrical Engineering from the University of Bristol, UK, in 2019. During her Ph.D. she was part of the PROVISION ITN project funded by the European Commission. The project was conducted by leading video compression research groups from academia and industry and focused on the development of perceptually optimized video coding. Her joint contribution on video compression using spatio-temporal resolution adaptation was awarded at the Grand Challenge on Video Compression Technology at ICIP 2017. Mariana is currently a Senior Research Scientist in the Encoding Technologies team at Netflix, which is responsible for developing and maintaining the encoding pipeline and producing all multimedia streams served to members around the globe. Furthermore, as a member of the Video Codecs & Research group, she works on a variety of R&D projects on video encoding optimization, video codec evaluation, quality metrics and the development and research for the next generation of video codecs.