[Colloquium] Reminder: Talks at TTI-C (Ramanan 3/31 and DeVore 4/1) THIS WEEK @ 3pm

[Katherine Cumming]

Toyota Technological Institute at Chicago Talk
 
Guest Speaker
 
Speaker:  Deva Ramanan, UC Berkeley
Speaker's homepage:  http://www.cs.berkeley.edu/~ramanan/
 
Time:  Thursday, March 31st @ 3:00pm
Location:  TTI-C Conference Room
Refreshments provided
 
Title:  Tracking People and Recognizing Their Activities
 
 
Abstract:  
An important, open vision problem is to automatically describe what
people are doing in a sequence of video. This problem is difficult for
several reasons. First, one needs to determine how many people (if any)
are in each frame and estimate their configurations (where they are and
what their arms and legs are doing). But finding people and localizing
their limbs is hard because people (a) wear a variety of different
clothes, (b) appear in a variety of poses and (c) tend to partially
occlude themselves and each other. Secondly, one must sew together
estimated configuration reports from across frames into a motion path;
this is tricky because people can move fast and unpredictably. Finally,
one must describe what each person is doing; this problem is poorly
understood, not least because there is no natural or canonical set of
categories into which to classify activities. In this talk I will
discuss our progress on this problem. We develop a tracker that works in
two stages; it first (a) builds a model of appearance of each person in
a video and then (b) tracks by detecting those models in each frame
("tracking by model-building and detection"). We then marry our tracker
with a motion synthesis engine that works by re-assembling pre-recorded
motion clips. The synthesis engine generates new motions that are
human-like and close to the image measurements reported by the tracker.
By using labeled motion clips, our synthesizer also generates activity
labels for each image frame ("analysis by synthesis"). We have
extensively tested our system, running it on hundreds of thousands of
frames of unscripted indoor and outdoor activity, a feature-length film,
and legacy sports footage. 
 
-----------------------------------------------------------------
 
 
Toyota Technological Institute at Chicago Talk
 
Learning Theory Program
Speaker:  Ronald A. DeVore, University of South Carolina
Speaker's homepage:   <http://www.math.sc.edu/~devore/>
http://www.math.sc.edu/~devore/
Time:  Friday, April 1st @ 3:00pm
Title:  The Regression Problem in Supervised Learning
Abstract:
We shall give an overview of the regression problem in supervised
learning. Our goal is to understand what is the best that we can expect
from learning algorithms in terms of accuracy performance in either
expectation or probability. 

We shall draw a contrast between (i) prior dependent algorithms where an
assumption on the prior is used in the algorithm; and (ii) universal
algorithms (called model selection in statistics) in which no knowledge
of a prior is assumed in the algorithm itself. Of course, the evaluation
of the performance of the algorithm can only be made under some prior
conditions. We shall consider such algorithm evaluation under priors
described by either smoothness of the regression function or by some
assumption on its approximability.

This presentation will provide an overview of the collaborative research
of the speaker with Gerard Kerkyacharian, Dominique Picard, and Vladimir
Temlyakov. It is also a launching point for the construction of
numerically friendly adaptive algorithms for learning recently
constructed by the speaker and Peter Binev, Albert Cohen, Wolfgang
Dahmen, and Vladimir Temlyakov. 
If you have questions, or would like to meet the speaker, please contact
Katherine at 4-1994 or kcumming at tti-c.org. For information on future
TTI-C talks or events, please go to the TTI-C Events
<http://ttic.uchicago.edu/events/events_dyn.php>  page. 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20050329/2c0a981b/attachment.htm