Professor Mann's research interests are in computational perception with emphasis on motion perception and event recognition. Mann's long term research goal is to develop a rigorous computational framework for visual perception that leads to effective tools for tackling a wide range of problems in computational vision. His research seeks to answer two fundamental questions in perception: First, what computational machinery is necessary to build an artificial (computational) perceptual system. Second, what type of world knowledge is necessary for a perceptual system to "understand'' a particular visual domain.
A second branch of Mann's research addresses the general machinery needed for perception. Perception can be viewed as a process of "unconscious inference'' in which the perceiver finds interpretation(s) of the world that best explain its observations. To understand this process Mann is exploring the connection between perception and Bayesian inference. He is currently applying Bayesian approaches to image segmentation and grouping, object recognition, and motion understanding.
Mann's most recent research results are in pen-based input systems, gesture input for human computer interaction, and Bayesian methods for image alignment.
Degrees and awards
BEng, MEng (McMaster), PhD (Toronto)
Industrial and sabbatical experience
Before coming to Waterloo, Mann did postdoctoral work (with Jeff Siskind) at NEC Research Institute, in Princeton, New Jersey. After completing his master's degree, Mann was a visiting researcher (with Teuvo Kohonen) in Helsinki, Finland.
A. Fourney, M. Terry, and R. Mann. Gesturing in the Wild: Understanding the Effects and Implications of Gesture-Based Interaction for Dynamic Presentations. 24th British Computer Society Conference on Human Computer Interaction. Dundee, Scotland. September 2010.
J. Orchard and R. Mann. Registering a Multi-Sensor Ensemble of Images. IEEE Transactions on Image Processing 19(5):1235-1247, May 2010.
N. Miller and R. Mann. Detecting Hand-Ball Events in Video Sequences. Proceedings of Canadian Conference on Computer and Robot Vision (CRV), 2008.
J. Ruiz, D. Tausky, A. Bunt, E. Lank, and R. Mann. Analyzing the kinematics of bivariate pointing. Proceedings of Graphics Interface (GI), 2008.
R. Mann and M. Langer. Spectral estimation of motion parallax and application to egomotion. Journal of the Optical Society of America A, 22(9):1717-1731, 2005.
D. Tausky and R. Mann. Categorization and Learning of Pen Motion Using Hidden Markov Models. Proceedings of Canadian Conference on Computer and Robot Vision (CRV), 2004.
R. Mann and A. Jepson. Detection and Classification of Motion Boundaries. Proceedings of Eighteenth National Conference on Artificial Intelligence (AAAI), 2002.