Senior Software Engineer, Algorithms and Research
Zhi Li has been with the Video Algorithms team at Netflix since November 2014. He has been leading the VMAF project over the past three years. Before VMAF, he was the main designer of Netflix’s first-generation Per-title Encoding architecture. Before joining Netflix, he worked at Cisco as a research engineer, focusing on ABR adaptive streaming algorithms. He obtained his B.Eng and M.Eng degrees from National University of Singapore, and Ph.D. degree from Stanford University. He received two best paper awards for his work on multimedia security. He has published dozens of journal/conference papers and authored dozens of international patents.
VMAF (Video Multi-Assessment Fusion) is a quality metric that combines human vision modeling with machine learning. It demonstrates high correlation to human perception and gives a score that is consistent across content. VMAF has been widely applied at Netflix in areas such as video quality monitoring and encoding optimization. VMAF was released on Github in 2016 and has had considerable updates since that time. This talk focuses on the latest VMAF improvements and enrichments, such as speed optimization, accurate models to predict mobile and 4K TV viewing conditions, and adding a confidence interval to quantify the level of confidence in the quality prediction. In addition, we discuss VMAF use cases and look at the VMAF road map for the near future.