Recent work from the team includes improving YouTube thumbnails with deep neural nets, the YouTube-8M dataset for video understanding research, Motion Stills apps on Android and iOS for machine intelligence-enabled microvideo creation, and learned image super-resolution (RAISR).
Tomáš came to Google by way of the Massachusetts Institute of Technology, where he received a Ph.D. in computer science in 2007. His doctoral research focused on motion tracking, scene activity analysis and visual attention models. Prior to attending MIT, Tomas received a B.S. in computer science from Yale University, where he did undergraduate research in computational models of visual cortical processing.