Xiangyang Lan

Upson Hall Room 4156, Cornell University, Ithaca, NY, 14850

Phone: 607-255-5033 (office), 607-342-4542 (cell)

Homepage: http://www.cs.cornell.edu/~xylan

Email: xylan@cs.cornell.edu

Objective

 

To find a permanent position in research labs, research-oriented companies, or financial corporations.

 

Research Interest

 

Probabilistic Modeling, Machine Learning, with application to Computer Vision, Data Mining, and Financial Modeling.

 

Education

 

  • Cornell University, Ithaca, NY

GPA: 4.00/4.00

09/2002 – present

Ph.D. Candidate in Computer Science (expected 06/2007)

Advisor: Prof. Dan Huttenlocher

 

  • Tsinghua University, Beijing, China

GPA: 90.4/100

09/1998 – 07/2002

B. E. in Computer Science and Technology.

 

Qualifications

 

  • Advanced knowledge in computer vision, image processing, and machine learning.
  • Advanced knowledge in graphical models, probabilistic modeling, statistical inference and learning.
  • Strong algorithm design and analysis ability. Proficient in standard data structures.
  • Proficient in C/C++/STL, Matlab, Java. Programming experience with Perl, Python, scripts.
  • Experienced with small-scale system building and debugging.

 

Research Experience

 

  • Research Assistant, Computer Science Dept., Cornell University

Advisor: Prof. Dan Huttenlocher

Ithaca, NY

09/02 – present

o        Visual modeling and analysis of deformable objects.

o        Visual analysis of articulated body and human tracking against view change and occlusion.

o        Efficient inference of higher order Markov Random Field using Belief Propagation.

 

  • Research Assistant, Computer Science Dept., Cornell University

Advisor: Prof. Dan Huttenlocher , Prof. Jon Kleinberg

Ithaca, NY

01/06 – present

o        Data mining on large-scale on-line social network to study the evolution of online communities.

o        Multiple time series correlation analysis based on burst detection and alignment.

 

  • Intern Researcher, IVM Group, Microsoft Research, Microsoft Corp.

Redmond, WA

Advisor: Dr. Larry Zitnick

05/06 – 08/06

o        Proposed a novel probabilistic model for feature-based object recognition. Our model is not an image-centric approach, and can handle large appearance variation and global deformation.

o        Patent filed. Submitted for publication.

 

  • Software Developer, Microsoft Research, Microsoft Corp.

Beijing, China

o        Developed a system for video summarization based on the modeling of viewer’s attention.

02/02 – 06/02

 

Publications

 

  • Lars Backstrom, Dan Huttenlocher, Jon Kleinberg, Xiangyang Lan.  

Group Formation in Large Social Networks: Membership, Growth, and Evolution.

Proc. 12th ACM SIGKDD Intl. Conf. on Knowledge Discovery and Data Mining, 2006.

 

  • Xiangyang Lan, Stefan Roth, Dan Huttenlocher, Michael Black.
    Efficient Belief Propagation with Learned Higher-Order Markov Random Fields.

In Proc. of European Conference on Computer Vision (ECCV), 2006.

 

  • Xiangyang Lan and Dan Huttenlocher.
    Beyond Trees: Common-Factor Models for 2D Human Pose Recovery.

In Proc. of IEEE Intl. Conf. on Computer Vision (ICCV), 2005.

 

  • Xiangyang Lan and Dan Huttenlocher.
    A Unified Spatio-Temporal Articulated Model for Tracking.

In Proc. of Computer Vision and Pattern Recognition (CVPR), 2004.

 

Teaching Experience

 

  • Introduction to Computer Programming (CS 100), Spring 2004
  • Computers and Programming (CS 211), Fall 2003
  • Computer Networks (CS 519), Spring 2003
  • Introduction to Theory of Computing (CS 381), Fall 2002

02/04 – 05/04

09/03 – 12/03

02/03 – 05/03

09/02 – 12/02

Responsible for giving review sessions, design and grading of homework and project assignments.

Received T.A. Award by Computer Science Dept. Cornell University.

 

Patent

 

  • Local Bi-Gram Model for Object Recognition”. Co-inventor with Larry Zitnick and Richard Szeliski, US Patent pending, 2006

 

Presentations

 

  • Probabilistic Graphical Models for Object Recognition”, Ph.D. Dissertation proposal

Computer Science Department, Cornell University, Ithaca, NY, May, 2006

 

  • Triplet Tree for Object Recognition”, MSR Intern final presentation

Microsoft Research, Redmond, WA, August, 2006

 

  • Beyond Trees: Common-Factor Models for 2D Human Pose Recovery”, ICCV oral presentation

ICCV Conference, Beijing, China, Oct, 2005

 

Honors and Awards

 

  • T.A. Award for Excellence, Computer Science Department, Cornell University

04/2004

  • "12-9" Fellowship for Outstanding Student 2000-2001 in Tsinghua University.

11/2001

  • Meritorious Winner in the 2001 Mathematical Contest in Modeling (MCM'2001)

04/2001

  • Scholarship for Academic Excellence in Tsinghua University

09/2000

 

Other Skills

 

  • Software development:

Familiar with Software Development life cycle and Object Oriented design.

  • Finance Accounting:

Familiar with Financial Statement / Ratio Analysis and other basic financial concepts like CAPM and Capital Structure policy.

  • Language:

Fluent in English, Chinese, and Cantonese.