ABSTRACT: This talk will give an introduction to what the Computer Vision research is focused on in Facebook AI organization. Facebook AI has close to 500 researchers and engineers distributed across Menlo Park-CA, New York City , Boston, Canada and Europe. Our mission in computer vision is to conduct cutting-edge fundamental research and solve computer vision problems that are important to Facebook’s missions and products. Our research spans across video analysis, OCR, person identification, adversarial learning, transfer learning, and extreme vision and we publish in top computer vision conferences as we make progress on these problems. At the same time, we have a productization-focused team that works on best engineering practices and infrastructure in order to take our technologies to product teams that will greatly benefit. BIO: Matt Feiszli, Research Scientist Manager, leads computer vision efforts on video understanding at Facebook. He was previously leading video understanding as Senior Machine Learning Scientist at Sentient Technologies and also spent four years at Yale University as Gibbs Assistant Professor in mathematics. He received his Ph.D. of Applied Mathematics from Brown University with David Mumford and has undergraduate degrees in Computer Science and Psychology from Yale University.

Hosted by Professor Milos Ercegovac

REFRESHMENTS at 3:45 pm, SPEAKER at 4:15 pm