Abstract:  Predicting the perceptual quality of pictures and videos is a hard problem that has been successfully addressed in many scenarios, such as quality control of streaming videos and sharing of social media pictures. In this talk I will address how visual quality perception can be understood using principles of neuroscience and neuro-statistical models of distortion. In particular I will review some basic vision science that makes accurate perceptual visual quality prediction possible, and how algorithms are designed that are now used worldwide. However, generated content like gaming videos and GenAI pictures are fundamentally different from optical pictures, and those differences may be statistically testable. I will discuss a recent approach to gaming content quality prediction called GAMIVAL that combines neuro-statistical models adapted to gaming distortions with deep features that capture gaming semantics. I will also discuss an early exploration into the perceptual statistics of GenAI pictures, and how these might be used to gauge GenAI visual quality and detectability.
 

Bio: Al Bovik is the Cockrell Family Regents Endowed Chair Professor at The University of Texas at Austin. His research interests land at the nexus of visual neuroscience and digital pictures and videos, and how their capture, processing, transmission, and display can be perceptually optimized. An elected member of the U.S. National Academy of Engineering, the Indian National Academy of Engineering, the National Academy of Inventors, and Academia Europaea, his many honors include the IEEE Edison Medal, a Primetime Emmy Award, a Technology & Engineering Emmy Award, the RPS Progress Medal, and the Edwin H. Land Medal.
http://utw10503.utweb.utexas.edu/bovik.php, and https://www.colorado.edu/ecee/al-bovik