About
This page presents some of my research work in Computer Vision and Image Analysis, much of it conducted at Palo Alto Research Center (PARC, formerly Xerox PARC). One main area of interest is Ullman's notion of Visual Routines -- its basis in computation structures and algorithms, its connection to perceptual organization and figure ground separation processes, and its application to machine perception of hand-drawn and printed graphical documents. Another main research area is recognition by constrained optimization, particularly as applied to hand drawn figures.
Contact
research@jamesmahoney.com 650 814 1734
Projects

Some principal projects are presented here in roughly chronological order. Links to papers, manuscripts, talks, or patents are provided in many cases.

This section is in an early stage of development. I'll be adding some discussion of the various topics soon.

Global Visual Salience
FG01ThreeProminentBlobs
My masters thesis was a broad examination of basic operations for Visual Routines, but some of its most notable observations were about the phenomenon of global visual salience.
Image Chunking: Defining Spatial Building Blocks for Scene Analysis.
James V. Mahoney.
M.I.T. Artificial Intelligence Laboratory Technical Report AITR-980, 1987. pdf
“Image Chunking: Defining Spatial Building Blocks for Scene Analysis.”
James V. Mahoney and Shimon Ullman.
In Computational Processes in Human Vision: An Interdisciplinary Perspective, Zenon Pylyshyn (Ed.) Canadian Institute for Advanced Research Series in Artificial Intelligence, 1989.
Signal-based Scene Analysis
MahoneyEH-TR_03b
Position-invariant/Exhaustive Hierarchical Image Analysis; Signal-Based Visual Routines.
Image Chunking.
James V. Mahoney.
Xerox Palo Alto Research Center, Blue-and-White Series Technical Report, April 1992.
Image Linking.
James V. Mahoney.
Xerox Palo Alto Research Center, Blue-and-White Series Technical Report, May 1992.
Position-invariant Hierarchical Image Analysis.
James V. Mahoney.
Internal memo, 1994. pdf
Signal-based Visual Routines.
James V. Mahoney.
Internal memo, 1995 pdf
Functional Visual Routines.
James V. Mahoney.
Internal memo, 1996
Detecting the running text in a page image.
James V. Mahoney.
Internal memo, 1996
How qualitative spatial reasoning can improve strategy game AIs.
Kenneth D. Forbus, James V. Mahoney, Kevin Dill.
IEEE Intelligent Systems, Volume 17, Issue 4 (July 2002) pdf
Recognizing Sketched Diagrams
Sketchalyzer1
PARC research on recognizing hand-drawn charts and graphs.
Sketchalyzer: a paper user interface for desiging and editing graphic documents.
Mahoney, James V.; Rao, Satyajit
Internal memo, 1994.
Analyzing an image or other data to obtain a stable number of groups.
Mahoney, James V.; Rao, Satyajit
U.S. Patent 5,537,491.
Analyzing an image showing a graphical representation of a layout.
Mahoney, James V.; Rao, Satyajit
U.S. Patent 5,455,898.
Recognizing Sloppy Line Drawings
StickManMatches
Matching sloppy stick figures.
Interpreting Sloppy Stick Figures by Graph Rectification and Constraint-based Matching.
James V. Mahoney and Markus P.J. Fromherz.
Lecture Notes in Computer Science, Vol. 2390, Blostein, Dorothea; Kwon, Young-Bin (Eds.): Springer-Verlag, 2002. pdf
Interpreting Sloppy Stick Figures with Constraint-based Subgraph Matching
James V. Mahoney and Markus P.J. Fromherz.
Constraint Programming 2001, Paphos, Cyprus, Nov. 2001. pdf
Three main concerns in sketch recognition and an approach to addressing them
James V. Mahoney and Markus P.J. Fromherz.
AAAI Spring Symposium on Sketch Understanding, pages 105--112, March 25-27 2002 pdf
Publications and Patents

A lot of my research was patented rather than published. Patents in this listing are indicated by italics.

Base Representations and Basic Operations for Visual Routines
James V. Mahoney. Image Chunking: Defining Spatial Building Blocks for Scene Analysis. M.I.T. Artificial Intelligence Laboratory Technical Report AITR-980, August, 1987.
James V. Mahoney and Shimon Ullman. “Image Chunking: Defining Spatial Building Blocks for Scene Analysis.” In Computational Processes in Human Vision: An Interdisciplinary Perspective, Zenon Pylyshyn (Ed.) Canadian Institute for Advanced Research Series in Artificial Intelligence, 1989.
James V. Mahoney. Image Links and Their Applications. Xerox Palo Alto Research Center, Blue-and-White Series Technical Report. April 1992.
James V. Mahoney. Image Chunks and Their Applications. Xerox Palo Alto Research Center, Blue-and-White Series Technical Report. May 1992.
Mahoney, James V.; Clemens; David T., Exhaustive hierarchical near neighbor operations on an image. U.S. Patent 5,305,395.
Mahoney, James V., Dense aggregative hierarhical techniques for data analysis. U.S. Patent 5,280,547.
Mahoney, James V., Comparison of image shapes based on near neighbor data. U.S. Patent 5,255,354.
Mahoney, James V., Labeling pixels of an image based on near neighbor attributes. U.S. Patent 5,239,596.
Mahoney, James V., Hierarchical operations on border attribute data for image regions. U.S. Patent 5,231,676.
Mahoney, James V., Local hierarchical processing focus shift within an image. U.S. Patent 5,193,125.
Diagram and Layout Understanding Applications of Visual Routines
Mahoney, James V., Method for classifying non-running text in an image. U.S. Patent 6,009,196.
Mahoney, James V., Method and apparatus for detecting running text in an image. U.S. Patent 5,889,886.
Mahoney, James V.; Rao, Satyajit, Analyzing an image showing editing marks to obtain category of editing operation. U.S. Patent 5,659,639.
Mahoney, James V., Using an image showing a perimeter relationship representation to obtain data indicating a relationship among distinctions. U.S. Patent 5,563,991.
Mahoney, James V.; Rao, Satyajit, Using a category to analyze an image showing a graphical representation. U.S. Patent 5,544,267.
Mahoney, James V.; Rao, Satyajit, Analyzing an image or other data to obtain a stable number of groups. U.S. Patent 5,537,491.
Rao, Satyajit ; Mahoney, James V., Analyzing an image showing a node-link structure. U.S. Patent 5,522,022.
Rao, Satyajit ; Mahoney, James V., Analyzing an image showing a proportioned parts graph. U.S. Patent 5,513,271.
Mahoney, James V.; Rao, Satyajit, Analyzing an image showing a graphical representation of a layout. U.S. Patent 5,455,898.
Mahoney, James V., Analyzing an image showing a row/column representation. U.S. Patent 5,392,130.
Document Retrieval Systems
Mahoney, James V.; Bobrow, Daniel G.; Rucklidge, William J., Transmission of subsets of layout objects at different resolutions. U.S. Patent 6,665,841.
Bobrow, Daniel G.; Mahoney, James V.; Rucklidge, William J., Sorting image segments into clusters based on a distance measurement. U.S. Patent 6,562,077.
Mahoney, James V.; Rucklidge, William J., System for performing collective symbol-based compression of a corpus of document images. U.S. Patent 6,020,972.
Mahoney, James V.; Blomberg, Jeanette L.; Trigg, Randall H.; Shin, Christian K., System for searching a corpus of document images by user specified document layout components. U.S. Patent 5,999,664.
Line drawing recognition by constrained optimization
James V. Mahoney and Markus P.J. Fromherz. Three Main Concerns in Sketch Recognition and an Approach to Addressing Them. AAAI Spring Symposium on Sketch Understanding, Stanford, CA, March 2002.
James V. Mahoney and Markus P.J. Fromherz. Handling Ambiguity in Constraint-based Recognition of Stick Figure Sketches. SPIE Document Recognition and Retrieval IX Conf., San Jose, CA, Jan. 2002.
Markus P.J. Fromherz and James V. Mahoney. Interpreting Sloppy Stick Figures with Constraint-based Subgraph Matching. Seventh Int. Conf. on Principles and Practice of Constraint Programming, Paphos, Cyprus, Nov. 2001.
James V. Mahoney and Markus P.J. Fromherz. Interpreting Sloppy Stick Figures by Graph Rectification and Constraint-based Matching. Lecture Notes in Computer Science, Vol. 2390, Blostein, Dorothea; Kwon, Young-Bin (Eds.): Springer-Verlag, 2002.
James V. Mahoney and Markus P.J. Fromherz. Perceptual Organization as Graph Rectification in a Constraint-based Scheme for Interpreting Sloppy Stick Figures (Extended Abstract). Perceptual Organization in Computer Vision '01, Canada, July 2001.
Perceptually-suported document editing and user interfaces
Saund, E.; Mahoney, J. V. Perceptual support of diagram creation and editing. In A. Blackwell, K. Marriott, A. Shimojima, editors. Diagrammatic Representation and Inference; Third International Conference on the Theory and Applications of Diagrams (Diagrams 2004); 2004 March 22-24; Cambridge; UK. Springer LNAI 2980: 424-427.
Saund, E.; Mahoney, J. V. ScanScribe: perceptually supported diagram image editing. In A. Blackwell, K. Marriott, A. Shimojima, editors. Diagrammatic Representation and Inference; Proceedings of the Third International Conference on the Theory and Application of Diagrams (Diagrams 2004); 2004 March 22-24; Cambridge; UK. Springer LNAI 2980: 428-431.
Saund, E.; Fleet, D. J.; Mahoney, J. V.; Larner, D. L. Rough and degraded document interpretation by perceptual organization. 2003 Symposium on Document Image Understanding Technology (SDIUT 03); 2003 April 9-11; Greenbelt; MD.
Saund, E.; Fleet, D. J.; Larner, D. L.; Mahoney, J. V. Perceptually-supported image editing of text and graphics. Proceedings of the 16th Annual ACM Symposium on User Interface Sofware and Technology (UIST 03); 2003 November 2-5; Vancouver; BC; Canada. NY: ACM; 2003; 183-192.
Saund, E.; Fleet, D. J.; Mahoney, J. V.; Larner, D. L. Pressing vision problems in intelligent sketch understanding systems. Bay Area Vision Meeting; 2003 June 11; Santa Cruz; CA.
Saund, E.; Mahoney, J. V.; Fleet, D. J.; Larner, D. L.; Lank, E. Perceptual organization as a foundation for intelligent sketch editing. 2002 AAAI Spring Symposium on Sketch Editing, TR SS-02-08; 2002 March 25 - 27; Palo Alto; CA; USA.
Saund, Eric ; Moran, Thomas P.; Larner, Daniel L.; Mahoney, James V.; Fleet, David J.; Popat, Ashok C., System and method for editing electronic images. U.S. Patent 6,903,751.
Mahoney, James V.; Chiu, Patrick ; Moran; Thomas P.; van Melle, William J, Automatic extraction of text regions and region borders for an electronic work surface. U.S. Patent 6,470,095.
Links
UNDER CONSTRUCTION