|
RESEARCH
INTERESTS
General interest in Artificial Intelligence, Machine Learning, and
Cognitive Science research problems. More specifically, I have
experience and interest in problems pertaining to Computer Vision and
Object Recognition.
EDUCATION
Ph.D. in Computer
Science
Carnegie Mellon University (expected
2010)
B.S. in Cognitive Science (Specialization in Computation) w/ Honors
Overall GPA: 3.7
Minor in Computer Science and Engineering
University of California, San Diego (June 2005)005)
SKILLS
Languages: Java, C/C++, Matlab, Visual Basic, LaTeX, HTML, XML, Flash,
Perl, CSH and Bash Script. Applications: Eclipse, Microsoft .Net,
MS Office, Tekkotsu Framework, Photoshop, Flash. Other languages and
applications also, feel free to contact with additional questions.
EXPERIENCE
Research Assistant, Carnegie Mellon University
(September 2005 - Present)
Advisor: Manuela Veloso
Investigating an approach to integrating multiple cues of diverse types
(visual, activity, speech, etc.) for more robust object recognition
using video data and probabilistic relational learning techniques.
Internship, Lockheed Martin
(June 2009 –
August 2009)
Manager: Mark Gersh
Supervisor: Andy Zimdars, Dave Tyler
Two projects: (1) Developed vision and assessed feasibility of visual
navigation, feature recognition and obstacle avoidance with the robots'
limited on-board processors for DARPA LANdroids project (2) Image
registration and its Cramer-Rao lower bound for spectral images. Please
contact me for more details.
Teaching assistant, Carnegie Mellon University
(January 2009 – May 2009)
Assisting in “Computer Vision” course with Prof. Tai Sing Lee including
typical Computer Vision routines, explaining basic principles and
techniques to students, including running sessions and giving one
lecture.
Teaching assistant, Carnegie Mellon University
(January 2007 – May 2007)
Assisting in “Cognitive Robotics”, a course with David Touretzky which
teaches robot programming with high level primitives for perception and
action based on cognitive science concepts such as visual routines, dual
coding theory, and affordances using the Tekkotsu software framework and
the Sony Aibo robot dog.
Honors Project, UC San Diego
(September 2004 – June 2005)
Developed a computer model of the use
of saccades in developing a position-invariant representation in object
recognition.
Independent Research in Cognitive Science
(January 2004 – June 2005)
Assisted in computer vision project researching object recognition using
shared-features.
Other Courses and Studies
(September 2001 – Present)
Familiar with concepts and techniques in other AI and cognitive science
subjects such as Ethnography, Computer User Design, Genetic Programming,
Path Planning.
HONORS/AWARDS
Women @ IT fellowship
(September 2005 – June 2006)
Cognitive Science Honors
(June 2005)
Gates Millennium Scholarship
(June 2001-2005)
PUBLICATIONS
Aboutalib, S. and
Veloso,M. Multiple-Cue Object Recognition on Outside Datasets. IEEE/RSJ
International Conference on Intelligent Robots and Systems (IROS),
2010. Under Submission.
Aboutalib, S. and
Veloso,M. Cue-based equivalence classes and incremental
discrimination for object recognition. IEEE/RSJ
International Conference on Intelligent Robots and Systems (IROS),
2009. [ Invited Speaker, ICVW ]
Aboutalib, S. and Veloso, M.
Simulation and Weights of Multiple
Cues for Robust Object Recognition.
In Proceedings of the IEEE
International Conference on Intelligent Robots and Systems (IROS),
San Diego, CA (October 2007)
Aboutalib, S. and Veloso, M. Towards Using Multiple Cues for Robust
Object Recognition. In Proceedings of AAMAS'07, the Sixth
International Joint Conference on Autonomous Agents and Multi-Agent
Systems, Honolulu, HI (May 2007)
Murphy-Chutorian, E., Aboutalib, S., Triesch, J. Analysis of a
Biologically-Inspired System for Real-time Object Recognition
Cognitive Science Online, 3.2, pp. 1-14. http://cogsci-online.ucsd.edu/3/3-3.pdf
(2005)
* Thesis proposal document available upon request.
|