Ankur Gupta
Doctoral Student
Department of Computer Science
The University of British Columbia

Email: ankgupta AT cs dot ubc.ca
Supervisors: Bob Woodham and Jim Little
Lab: Laboratory of Computational Intelligence (LCI)
X210, ICICS/CS Building

Short Bio
I am a PhD student in the Computer Science department at UBC.


Computer Vision & Robotics are my areas of research. Currently, I am interested in motion tracking and active vision.

I completed my undergraduate degree in Electrical Engineering from the Indian Institute of Technology, Kanpur, India.

I enjoy running, hiking, and taking pictures.
 

Publications

Conference papers

Ankur Gupta, James J. Little, Robert J. Woodham. "Using Line and Ellipse Features for Rectification of Broadcast Hockey Video". Canadian Conference on Computer and Robot Vision, 2011.
[pdf] [Project page]


David Meger, Marius Muja, Scott Helmer, Ankur Gupta, Catherine Gamroth, Tomas Hoffman, Matthew Baumann, Tristram Southey, Pooyan Fazli, Walter Wohlkinger, Pooja Viswanathan, James J. Little, David G. Lowe, and James Orwell. "Curious George: An Integrated Visual Search Platform". Canadian Conference on Computer and Robot Vision, 2010. 
[pdf] [Project page]

D. Meger, A. Gupta, and J. J. Little. "Viewpoint Detection Models for Sequential Embodied Object Category Recognition". International Conference on Robotics and Automation, 2010.
[pdf] [Project page]

Master's Thesis

Ankur Gupta, "Using Line and Ellipse Features for Rectification of Broadcast Hockey Video", Master's thesis,     The University of British Columbia, 2010.

[pdf] [Project page]


Projects


Image Sample
Automated sports video analysis

My master's thesis was a part of this project. I worked on  automatic rectification of broadcast quality hockey video sequences. For a moving camera case, this involves homography estimation between each frame and the geometric model of the rink. We show that, using ellipse and line features along with key-point based features,  the homography can be robustly estimated for a long video sequence.

[Output video] [A player tracking application using the output (video)]


Image Sample

Curious George project

Curious George is a semantic robot, capable of automatic navigation and object recognition (for both specific and generic objects). This was also UBC's entry to the Semantic Robot Vision Challenge (SRVC). Our team won the competition (software league) in 2009.

I also worked with David Meger and Jim Little on viewpoint planning for sequential object category recognition problem. This work was presented at ICRA, 2010.


Service

Currently, I serve as a:
- Computer Science Graduate Seminar Organizer. Here is the schedule for Grad Refresher 2011.
- Member of  Graduate Affairs Committee in the department.

20010-11
VP Administration/Finance, Computer Science Graduate Student Association (CSGSA)
Member, Graduate Affairs Committee, CSGSA

2009-10
Member, Orientation Committee, CSGSA
Mentor, Trimentoring Program
Mentor, Teaching Assistants Mentoring  Program

Links

Academic
Peter Kovesi's MATLAB functions for computer vision
MATLAB functions for multiple view geometry

Interesting blogs
Matt Might
Techcrunch
Willow Garage
Academic Productivity
Daniel Lemire's blog