If you have an event you would like posted, please contact


Loading Events

« All Events

  • This event has passed.

Seminar: Model-Based Perceptual Grouping and Shape Abstraction

April 8, 2015 @ 10:00 am - 12:00 pm

Sven Dickinson
Department of Computer Science
University of Toronto 

For many object classes, shape is the most generic feature for object categorization. However, when a strong shape prior, i.e., a target object, is not available, domain independent, mid-level shape priors must play a critical role in not only grouping causally related features, but regularizing or abstracting them to yield higher-order shape features that support object categorization. In this talk, I will present a framework in which mid-level shape priors take the form of a vocabulary of simple, user-defined 2-D part models.  From the vocabulary, we learn to not only group oversegmented regions into parts, but to abstract the shapes of the region groups, yielding a set of abstract part hypotheses.  However, the process of shape abstraction can be thought of as a form of “controlled hallucination”, which comes at the cost of many competing 2-D part hypotheses.  To improve part hypothesis precision, we assume that the 2-D parts represent the component faces of aspects that model a vocabulary of 3-D part models. We then exploit the relational structure (spatial context) of the faces encoded in the aspects, and again formulate hypothesis selection in a graph-theoretic, probabilistic framework.  Finally, we introduce a technique that is able to recover the pose and shape of a volumetric part from a recovered aspect, yielding a framework that revisits the classical problem of recovering a set of qualitative 3-D volumetric parts from a single 2-D image.

Sven Dickinson received the B.A.Sc. degree in Systems Design Engineering from the University of Waterloo, in 1983, and the M.S. and Ph.D. degrees in Computer Science from the University of Maryland, in 1988 and 1991, respectively. He is currently Professor and Chair of the Department of Computer Science at the University of Toronto. Prior to that, he was a faculty member at Rutgers University where he held a joint appointment between the Department of Computer Science and the Rutgers Center for Cognitive Science (RuCCS). His research focuses on object recognition, perceptual grouping, shape representation, and algorithms for inexact graph indexing and matching. He has published over 150 papers on these and other topics in computer vision, and currently serves or has served on the editorial boards of 9 journals.


April 8, 2015
10:00 am - 12:00 pm
Event Category:
Event Tags:


3033 Lassonde Building
York University
North York, Ontario M3J Canada
+ Google Map

Subscribe to our mailing list

Subscribed As