How should a robot perceive the world?
Speaker: Ashutosh Saxena, Cornell University
Date: Friday, November 16 2012
Time: 1:00PM to 2:00PM
Host: Stefanie Tellex, CSAIL
Contact: Stefanie Tellex, 617-383-4749, firstname.lastname@example.orgRelevant URL: http://www.cs.cornell.edu/~asaxena/
Title: How should a robot perceive the world?
Speaker: Ashutosh Saxena, Cornell University.
In order for a robot to perform tasks in the human environments, it
first needs to figure out "what" to perceive. While for some tasks,
just perceiving geometry and semantic labels is good enough, many
other tasks require a robot to be more creative. For example, for a
robot to arrange a disorganized room, it would need to perceive the
human preferences about the usage of objects as well as the low-level
In general, the key to successful perception lies in being able to
model the underlying "structure" in a problem. In this talk, I will
argue that the it is the humans that are the true underlying structure
in the problems. Therefore our algorithms should reason through
humans. This is not only true for tasks that involve humans explicitly
(such as human activity detection), but also true for tasks in which a
human was never observed!
I will demonstrate this idea through a few example robotic tasks,
including unloading items from a dishwasher, scene understanding,
loading a fridge, arranging a disorganized room, and performing
assistive tasks in response to human activities.
Ashutosh Saxena is an assistant professor in computer science
department at Cornell University. His research interests include
machine learning and robotics perception, especially in the domain of
personal robotics. He received his MS in 2006 and Ph.D. in 2009 from
Stanford University, and his B.Tech. in 2004 from Indian Institute of
Technology (IIT) Kanpur. He was a recipient of National Talent Scholar
award in India and Google Faculty award in 2011. He was also named a
Alfred P. Sloan Research Fellow in 2011 and a Microsoft Faculty Fellow
In the past, Ashutosh developed Make3D (http://make3d.cs.cornell.edu),
an algorithm that converts a single photograph into a 3D model. Tens of
thousands of users used this technology to convert their pictures to
3D. He has also developed algorithms that enable robots (such as
STAIR, POLAR, see http://pr.cs.cornell.edu) to perform household
chores such as unload items from a dishwasher, place items in a
fridge, etc. His work has received substantial amount of attention in
popular press, including the front-page of New York Times, BBC, ABC,
New Scientist, Discovery Science, and Wired Magazine. He has won best
paper awards in 3DRR and IEEE ACE, and was named a co-chair of the
IEEE technical committee on robot learning.
See other events happening in November 2012