Brown University
The Department of Computer Science Seminar
Refreshments will be served at 10:45 am |
Abstract: Analysis and detection of objects in images or actions in video sequences require a complex notion of similarity across visual data. Existing approaches are often based on extracting informative parameters or models learned from many prior examples of the visual data (object or action) of interest. These approaches, however, are often restricted to a small set of pre-defined classes of visual data, and do not generalize to scenarios with unfamiliar objects/actions. Moreover, in many realistic cases and problems, one has only a single example of the object or action of interest, or even NO explicit example whatsoever of what he is looking for...
In this talk I will show how we can infer about the global similarity of different complex visual data, by employing local similarities within and across these visual data. I will demonstrate the power of this approach through several example problems. These include: (i) Prediction of missing visual information in images and videos. (ii) Detection and retrieval of complex objects in cluttered images using a single example -- often only a rough hand-sketch of the object of interest. (iii) Detection of complex actions performed by differently dressed people against different backgrounds, based on a single example clip (without requiring fg/bg segmentation or motion estimation).
Joint work with Michal Irani (the first part also with Yonatan Wexler).
Host: Professor Michael Black
Brown University
Mathematics Department
Special Colloquium
Abstract: The group of invertible 2x2 matrices with integer coefficients has three different interpretations: it is the group of automorphisms of Z^{2}, the group of isotopy classes of diffeomorphisms of the 2-torus, and the group of outer automorphisms of the free group F_{2} of rank 2. These descriptions lead to three much studied families of discrete groups: GL_{n}(Z), mapping class groups, and Out(F_{n}). Even though the techniques differ, there is a kind of a dictionary between the basic results about these families and often a theorem in one class motivates the study in another. I will try to describe the basic spaces on which these groups act and parts of the dictionary.
Brown Analysis Seminar
PDE Seminar
Abstract: The solutions of Koch and Tataru are believed to provide an optimal framework for strong solutions of the Navier-Stokes equation. They correspond to data in BMO^{-1}. We will present results obtained in collaboration with N. Pavlovic and G. Staffilani that give estimates on the higher resgularity of these solutions. This has applications to questions of analyticity, decay, and study of the self-similar solutions.
<--- 2007 Index