Education

University of Southern California, Los Angeles, California, USA
Thesis Title: "Tensor Voting for Salient Feature Inference in Computer Vision"
Thesis Committee: Gerard Medioni (chair), Micheal Arbib, Irving Biederman

University of Hong Kong, Hong Kong
Thesis Title:"Linear-Time Motion Planning for Two Square, Movable Obstacles in a Grid Environment"

The Chinese University of Hong Kong, Hong Kong

Philips Research USA

Research and development of Computer Vision systems for professional and consumer applications, including:

• Hands-free videoconferencing for professional environments.
• Multimodal user interfaces for home videoconferencing and entertainment.
• Computer vision on silicon.
• Computer vision for ambient intelligent home.

Institute for Robotics and Intelligence Systems, University of Southern California

Derivation and implementation of a unified framework for the inference of multiple salient structures such as junctions, curves, regions, and surfaces from noisy, irregularly clustered data. This work is particularly useful in computer vi-sion as it tackles the common problems in different subfields of computer vision, that is, to extract salient and structured information from noisy and erroreous local measure-ments. My contributions are:

• Identification of the common subtasks in image understanding processes, which are, smooth region interpolation, discontinuity detection and outlier identification.
• Derivation of a tensorial framework for the inference of salient structures which tackles the 3 subtasks in image understanding simultaneously. The methodology is rooted in two elements: tensor calculus for data representation and non-linear vot-ing for data communication. The method is non-iterative, requires no initialization, and the only free parameter is scale.
• Application of the methodology to the inference of regions, curves, and junctions from sparse, noisy 2-D and 3-D data.
• Application of the methodology to the problems of shape from stereo, shape from shading and image sequence analysis.

Robotvis project, Institut National de Recherche en Informatique et en Automatique, France

Worked on the problem of computing structure and motion from a sparse set of views with Dr. R. Deriche and Dr. Z. Zhang of INRIA, France. Developed an al-gorithm for extracting geometry and motion information from a sparse set of images incrementally.

Department of Computer Science, University of Southern California

Responsible for conducting tutorials, grading assignments, and assisting students in the following courses: Operating System, Introduction to Programming Systems Design, Fundamentals of Computer Programming.

Department of Computer Science, University of Hong Kong

Responsible for conducting tutorials, grading assignments, and assisting students in the following courses: Introduction to data structures and algorithms, Compiling techniques, Computer net-works and data communications, and Design and analysis of algorithms.

Department of Computer Science, University of Hong Kong

Developed an linear algorithm for the analysis of a specific robot motion planning problem, in which the robot has to move objects in 2-D where the motion of the robot is restricted in a grid and the number of object is limited to two.

• Program Committee of IEEE Conference on Computer Vision and Pattern Recognition, 2001.
• Reviewer of IEEE Conference on Computer Vision and Pattern Recognition, 98.
• Reviewer of IEEE International Conference on Computer Vision, 95
• Reviewer of IEEE Conference on Computer Vision and Pattern Recognition, 95

• Undergraduate degree received with Second Class, Upper Division Honor
• College Deans List, December 1988

Computer Languages: C++, C, Lisp, Fortran, Pascal, Ada, Basic and Prolog
Systems: Windows NT/98/2000, MFC, Visual Studio, DirectX, Unix, MS DOS, X Window
Discipline: Computer Vision, Computer Graphics, Image Processing, Computational Geometry, Artificial Intelligence

• Gérard Medioni, Mi-Suen Lee and Chi-Keung Tang, A Computational Framework for Segmentation and Grouping, Elsevier Science B.V., Mar 2000.
• Mi-Suen Lee, Tensor Voting for Salient Feature Inference in Computer Vision, Ph.D. Dissertation, University of Southern California, 1998.

• C.-K. Tang, M.-S. Lee, G. Medioni, "Tensor Voting", in Perceptual Organization for Artificial Vision Systems, Kim L. Boyer and Sudeep Sarkar (Ed), Kluwer Academic Publishers, Boston , Mar 2000.

• Mi-Suen Lee, Gérard Medioni, and Philippos Mordohai, "Inference of Segmented, Overlapping Surfaces from Binocular Stereo", accepted to IEEE Transactions on Pattern Analysis and Machine Intelligence.
• Chi-Keung Tang, Gerard Medioni, and Mi-Suen Lee, "N-Dimensional Tensor Voting, and Application to Epipolar Geometry Estimation", in IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 23, no. 8, Aug 2001, pp.829-844.
• Mi-Suen Lee and Gérard Medioni, "Grouping .,-,->,O-, into Regions, Curves, and Junctions", Computer Vision and Image Understanding, vol. 76, no. 1, Oct 1999, pp. 54-69.
• Parag Havaldar, Mi-Suen Lee and Gérard Medioni, "Synthesizing Novel Views from Un-registered 2-D Images", Computer Graphics Forum, vol 16, no. 1, Mar 1997, pp. 65-74.

• Gérad Medioni, Chi-Keung Tang, and Mi-Suen Lee, "Tensor Voting: Theory and Applications", in 12eme Congres Francophone AFRIF-AFIA de Reconnaissance des Formes et Intelligence Artificielle (RFIA), Feb 2000.
• Mi-Suen Lee and Gérard Medioni, "Inferred Descriptions in terms of Curves, Regions and Junctions from Sparse, Noisy Binary Data", Proc. of International Workshop on Visual Forms, Capri, Italy, May 1997, pp. 350-367.

• Mi-Suen Lee, Daphna Weinshall, Eric Cohen-Solal, Antonio Colmenarez, and Damian Lyons, "A Computer Vision System for On-Screen Item Selection by Finger Pointing", accepted to IEEE Conference on Computer Vision andPattern Recognition, Hawaii, Dec 2001.
• Richard Kleihorst, Mi-Suen Lee, Anteneh A. Abbo, and Eric Cohen-Solal, "Real Time Skin-Region Detection with a Single-Chip Digital Camera", in Proc. IEEE Intl. Conf. Image Processing, Thessaloniki, Greece, Oct 2001.
• Philippos Mordohai, Gérard Medioni, and Mi-Suen Lee, "Inference of Segmented Overlapping Surfaces from Binocular and Multiple-View Stereo", in Proc. Third Workshop on Perceptual Organization in Computer Vision, July 2001.
• Hugo Strubbe and Mi-Suen Lee, "UI for a Videoconference Camera", in Proc. ACM CHI2001, Seattle, WA, Apr 2001, Extended Abstracts, pp. 333-334.
• Mi-Suen Lee, "Detecting People in Cluttered Indoor Scenes", in Proc. IEEE Conference on Computer Vision and Pattern Recognition, Hilton Head Island, SC, June 2000, pp. 804-809.
• Chi-Keung Tang, Gérard Medioni, and Mi-Suen Lee, "Epipolar Geometry Estimation by Tensor Voting in 8D", in Proc. International Conference on Computer Vision, Kerkyra, Greece, Sept, 1999, pp.502-509.
• Mircea Nicolescu, Gérard Medioni, and Mi-Suen Lee, "Segmentation, Tracking and Interpretation using Panoramic Video", in Proc. IEEE Workshop on Omnidirectional Vision, South Carolina, June 2000, pp.169-174.
• Mi-Suen Lee and Gérard Medioni, "Inferring Segmented Surface Description from Stereo Data", in Proc. Conference on Computer Vison and Pattern Recognition, Santa Barbara, CA, June, 1998, pp. 346-452.
• Mi-Suen Lee and Gérard Medioni, "A Unified Framework for Salient Curves, Regions, and Junctions Inference", Proc. 3rd Asian Conference on Computer Vision, Hong Kong, Jan.1998, pp. 315-322.
• Song Han, Mi-Suen Lee and Gérard Medioni, "Non-uniform Skew Estimation by Tensor Voting", Proc. IEEE Workshop on Document Image Analysis, San Juan, Puerto Rico, June 1997, pp. 1-4.
• Parag Havaldar, Mi-Suen Lee and Gérard Medioni, "View Synthesis from Unregistered 2- D Images", Proc. Graphics Interface'96, Toronto, Canada, May 1996, pp. 61-69.
• Mi-Suen Lee and Gérard Medioni, "Structure and Motion from a Sparse Set of Views", Proc. of Image Understanding Workshop, Palms Spring, CA, Feb 1996, pp. 1051-1056.
• Mi-Suen Lee and Gérard Medioni and R. Deriche, "Structure and Motion from a Sparse Set of Views", Proc. of IEEE Symposium on Computer Vision, Coral Gables, FL, Nov 1995, pp. 73-78.
• Mi-Suen Lee and Andrew Choi, "Linear-time Motion Planning for Two Square, Movable Obstacles in a Grid Environment", Proc. of IEEE Int. Workshop on Emerging Technologies and Factory Automation, Melbourne, Australia, Aug 1992