1998 - 1999 CS Annual Report                                                                  Faculty
choices.gif (4488 bytes)

Daniel P. Huttenlocher

Associate Professor
Cornell Weiss Presidential Fellow

PhD MIT, 1988

My main area of research is computer vision, and my recent work focuses on the problems of geometric shape compar- ison, model-based motion tracking and image segmentation. My approach to these problems involves a combination of theoretical analysis, algorithm design, implementation, and experimentation. The overall goal is to produce systems that work well in practice and have a firm underlying theoretical and algorithmic foundation. My group is developing vision 

dan.tif (84588 bytes)
systems for use in several application areas, including remote video surveillance and interactive document systems. Our work on interactive documents focuses on the use of images and video in learning and collaboration tools.   

University Activities 

  • Chair: University Faculty Advisory Board on Information Technology (FABIT) 
  • Chair: Provost's Task Force on Computing and Information Science  
  • Chair: Computer Science Facilities Committee  
  • Member: Project 2000 Executive Sponsor Committee  
  • Member: Weiss Fellow Award Committee  
  • Document-centered peer collaborations: An exploration of educational uses of networked communication technologies. Journal of Computer Mediated Communication 4, 3 (March
    1999) (with G. Gay, A.Sturgill, and W. Martin). 
  • A new bayesian approach to object recognition. Proceedings of the IEEE Computer Vision and Pattern Recognition Conference (1999) (with Y. Boykov). 
  • Fast detection of common geometric substructure in proteins. Proceedings of 3rd ACM RECOMB, Research in Computational Biology (Apr. 1999), 104-113 (with P. Chew, K. Kedem, and J. Kleinberg). 
  • Fontless structured document image representations for efficient rendering, U.S. Patent
    5,884,014, March, 1999 (with W. Rucklidge). 
  • Method and apparatus for  comparing symbols extracted from binary images of text using topology preserved dilated representations of the symbols, U.S. Patent 5,835,638, Nov, 1998 (with W. Rucklidge and E. Jaquith).