Matthew Mullin

College of Computing                            e-mail : mdmullin@cc.gatech.edu
Georgia Institue of Techonology                  Phone : (404) 385-4228
Atlanta, GA 30032-0280
U.S.A.

Research Interests
     Algorithms, Machine Learning, Computer Vision

Employment
     Georgia Institue of Technology                          2002­present

     Research Scientist I
       Research in computer vision and machine learning.

     Whizbang! Labs                                          2000­2002

     Distinguished Engineer
       Implemented machine learning algorithms for commercial text
       classification and extraction.

     Just Research (Justsystem Pittsburgh Research Center)   1998­2000

     Research Programmer
       Collaborated on projects in computer vision and machine learning
       resulting in publications and patents.

Educational Background

     Princeton University, Princeton, New Jersey             1986­1990
     Bachelor of Arts Degree with Honors in Mathematics

Journal Publications

     S.C. Brubaker, J. Wu, J. Sun, M.D. Mullin, and J.M. Rehg.
       On the design of cascades of boosted ensembles for face detection.
       International Journal of Computer Vision, 77(1­3):65­86, May 2008.
       DOI: 10.1007/s11263-007-0060-1.

     J. Wu, S.C. Brubaker, M.D. Mullin, and J.M. Rehg.
       Fast asymmetric learning for cascade face detection.
       IEEE Transactions on Pattern Recognition and Machine Intelligence,
       30(3):369­382, 2008. DOI: 10.1109/TPAMI.2007.1181.

     Z. Hussain, F. Laviolette, M. Marchand, J. Shawe-Taylor, S.C. Brubaker,
     and M.D. Mullin.
       Loss bounds for the set covering machine and sample-compression loss
       bounds for imbalanced data.
       Journal of Machine Learning Research, 8:2533­2549, 2007.

     J. Gallian and M. Mullin.
       Groups with anti-symmetric mappings.
       Archiv der Mathematik, 65(4):273­280, 1995. DOI: 10.1007/BF01195537.

Conference Publications

     S.C. Brubaker, J. Wu, J. Sun, M.D. Mullin, and J.M. Rehg.
       Towards optimal training of cascade classifiers.
       In Toward Category-Level Object Recognition, volume 4170 of Lecture
       Notes in Computer Science, pages 310­320. 2006. DOI: 10.1007/11957959-16.

     S.C. Brubaker, M.D. Mullin, and J.M. Rehg.
       Towards optimal training of cascade classifiers.
       In Proceedings of the European Conference on Computer Vision, 2006.

     J. Wu, M.D. Mullin, and J.M. Rehg.
       Linear asymmetric classifier for face detection.
       In Proceedings of the International Conference on Machine Learning, 2005.

     J. Wu, J.M. Rehg, and M.D. Mullin.
       Learning a rare event detection cascade by direct feature selection.
       In Proceedings of Neural Information Processing Systems, 2003.

     J. Wu, J.M. Rehg, and M.D. Mullin.
       Learning a rare event detection cascade by direct feature selection.
       In Workshop on Statistical and Computational Theories of Vision, 2003.

     R. Sukthankar, R.G. Stockton, and M.D. Mullin.
       Smarter presentations: Exploiting homography in camera-projector systems.
       In Proceedings of International Conference on Computer Vision, 2001.

     R. Sukthankar, R.G. Stockton, and M.D. Mullin.
       Self-calibrating camera-assisted presentation interface.
       In Proceedings of International Conference on Control, Automation,
       Robotics and Computer Vision, 2000.

     R. Sukthankar, R.G. Stockton, and Mullin M.D.
       Automatic keystone correction for camera-assisted presentation
       interfaces.
       In Proceedings of International Conference on Mulitmedia Interfaces, 2000

     M. Mullin and R. Sukthankar.
       Complete cross-validation for nearest neighbor classifiers.
       In Proceedings of International Conference on Machine Learning, 2000.

     T. Sim, R. Sukthankar, M. Mullin, and S. Baluja.
       Memory-based face recognition for visitor identification.
       In Proceedings of IEEE Face and Gesture, 2000.

     R. Caruana and M. Mullin.
       Estimating the number of local minima in big, nasty search spaces.
       In IJCAI99 Workshop on Statistical Machine Learning for Large-Scale
       Optimization, 1999.

Technical Reports

     J. Wu, J.M. Rehg, and M.D. Mullin.
       Learning a rare event detection cascade by direct feature selection.
       Technical report, Georgia Institute of Technology, 2003.

     T. Sim, R. Sukthankar, M.D. Mullin, and S. Baluja.
       High-performance memory-based face recognition for visitor identification
       Technical report, Just Research, 1999.

     M. Mullin and R. Sukthankar.
       An efficient technique for calculating exact nearest-neighbor
       classification accuracy.
       Technical report, Just Research, 1999.

Patents

     R. Sukthankar, R. Stockton, and M. Mullin.
       Method and apparatus for automatic keystone correction.
       U.S.  Patent 6,753,907, June 22, 2004.

     R. Sukthankar, R. Caruana, M. Mullin, and K. Kasegawa.
       Method and apparatus for using illumination from a display for computer
       vision based user interfaces and biometric authentication.
       U.S. Patent 6,704,447, March 9, 2004.

     R. Sukthankar, R. Stockton, and M. Mullin.
       Method and apparatus for calibrating projector-camera system.
       U.S. Patent 6,618,076, September 9, 2003.

Professional Activities
     Member of Association for Computing Machinery

     Member of the Mathematical Association of America