Title: Facial Analysis: Biometrics and Beyond
The fundamental goal of computer vision is to extract semantic information of objects from images and videos, where human face is one of the most studied objects. Given a facial image, automatic facial analysis is interested in answering questions such as, who is this person, where is he looking at, and what is his emotion. The technology development in facial analysis is making contributions to many real-world applications, including biometrics, surveillance, affective computing, HCI, medical, etc. This talk will walk you through some of my efforts in this area. I will start by introducing the thrust of research in face alignment via learning based approaches, which aims to accurately and efficiently fit a deformable landmark-based mesh to a face image and is one of the core capabilities to many facial analysis technologies. Then I will present a wide range of applications in the area of facial analysis.
Dr. Xiaoming Liu is a research scientist at General Electric (GE) Global Research. He received the B.E. degree from Beijing Information Technology Institute, China and the M.E. degree from Zhejiang University, China, in 1997 and 2000 respectively, both in Computer Science, and the Ph.D. degree in Electrical and Computer Engineering from Carnegie Mellon University in 2004. His research interests include computer vision, patter recognition, machine learning, biometrics, medical image analysis, etc. At GE, he performed research, and led a team of researchers on projects sponsored by various government agencies and GE businesses, aiming at developing computer vision algorithms to support facial and body analysis, multi-camera video surveillance systems, and medical imaging. He has actively directed a number of government-sponsored research programs totaling over $2 million with the role of project leader and principal investigator. He has delivered an array of technologies to GE businesses, including emotion analysis and NFL gaze estimation for NBCU, Biometrics ID kiosk for GE Security, Ultrasound scan plane identification for GE Healthcare, etc. He has authored more than 60 scientific publications with over 1000 citations and H-index 18, and has fired 20 U.S. patents.