Department of Computer Science
I am interested in visual learning, especially inducing properties of
the environment, i.e. what are relevant features? what is an
appropriate model of reasoning for a visual task? Typically this
involves bringing many sources of information to bear on a problem and
using them in a unified fashion.
More details may be found in my research projects or publications
Frequently Asked Questions
- Released the Matlab Grid CRF toolbox under the GNU General Public License.
- Paper "Teaching Computing as Science in a Research Experience" accepted for presentation at the 2015 SIGCSE Symposium (SIGCSE) [slides] [doi].
- Manuscript "Toward Integrated Scene Text Reading" published in the February 2014 IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI). [paper]
- Honored as an "Outstanding Reviewer" for CVPR 2013
- Paper "Toponym Recognition in Historical Maps by Gazetteer Alignment" accepted for presentation at the IAPR Intl. Conf. on Document Analysis and Recognition (ICDAR) 2013. [paper] [slides]
- Paper "Building Knowledge and Confidence with Mediascripting: A Successful Interdisciplinary Approach to CS1" presented at the 2013 SIGCSE Symposium. [paper]
- Gave invited talks on "Teaching Computers to Read" at Skidmore College, Mount Holyoke College, Smith College, and Colby College.
- Hosted the Birds-of-a-Feather session, "Imaging College Educators" and co-presented a poster "MediaScripting: teaching introductory CS through interactive graphics scripting" at the 2012 SIGCSE Symposium.
- Released the Matlab factor graph toolbox under the GNU General Public License.
- Grinnell College Department of Computer Science selected as a 2011-2012 CUDA Teaching Center.
- Gave an invited talk, "Robust Scene Text Recognition," at the
2011 University of Iowa Computing Conference.
- Manuscript "On Learning Conditional Random Fields for Stereo:
Exploring Model Structures and Approximate Inference" accepted for
publication in the International Journal of Computer Vision (IJCV). [paper]
- Chapter "Large-Scale Machine Learning" included in
the book GPU Gems Emerald Edition (Morgan Kaufmann). [paper]
- Paper "Typographical Features for Scene Text Recognition" accepted for publication at the International Conference on Pattern Recognition (ICPR) 2010. [paper] [slides]
- Hosted two Birds-of-a-Feather sessions at the 2010 SIGCSE Symposium, "Imaging College Educators" and "Teaching Social and Ethical Issues in CS0 and Beyond."
- Awarded an NVIDIA Professor Partnership Grant for proposal "GPU Computing in the Classroom and Directed Undergraduate Research."
- Paper "Scene Text Reading Using Similarity and a Lexicon with Sparse Belief Propagation" accepted for publication in the IEEE Transactions on Pattern Analysis and Machine Analysis (PAMI) special issue on Graphical Models in Computer Vision. [paper]
- Paper "A Discriminative Semi-Markov Model for Robust Scene Text Recognition" accepted for publication at the International Conference on Pattern Recognition (ICPR) 2008. [paper] [poster]
- Paper "Efficiently Learning Random Fields for Stereo Vision with Sparse Message Passing" accepted for publication at the European Conference on Computer Vision (ECCV) 2008. [paper] [poster]
- Gave an invited talk at University of Rochester on probabilistic models for scene text recognition.
- Presented a paper and talk at NESCAI 2008 on robust scene text recognition [paper], [slides], [poster].
Defended my dissertation on "Unified Detection and Recognition for Reading Text in Scene Images." Here is the final document, and the seminar slides.
Presented a talk at ICDAR 2007 on fast lexicon integration for scene text recognition. [slides]
Presented a session on "Leading Discussions in Science Labs" at the UMass Center for Teaching's TA orientation. [slides]
|Department of Computer Science|
|1116 8th Ave|
|Grinnell, IA 50112|
|GPG Public Key:||[Binary] [ASCII]|
|Office:|| Noyce '49 Science Center 3825|