Chris Sweeney

cmsweeney@cs.ucsb.edu

About Me

I am currently a fourth year graduate student at University of California, Santa Barbara pursuing a Ph.D. in Computer Science with a focus on Computer Vision, where I am a member of the Four Eyes Lab. I received my B.S. in Computer Science with high distinction from the University of Virginia in May 2011 as an undergraduate in the UVa Computer Graphics Group.

My current research focuses on making large-scale Structure from Motion (SfM) more accessible by removing constraints (e.g., requiring camera calibration) and common assumptions (e.g., assuming perspective cameras). I am currently funded by a 3 year NSF Graduate Research Fellowship.

2014-2015  Visiting student at ETH Zurich
2011-Now   PhD Student at UCSB
2013   Received 3 year NSF Fellowship
2012, 2013   Intern for Google Goggles
2011   B.S. in Math and Computer Science, UVA

More information can be found on my C. V.

Publications

My Google Scholar page



Efficient Computation of Absolute Pose for Gravity-Aware Augmented Reality [code]
Chris Sweeney, John Flynn, Benjamin Nuernberger, Matthew Turk, Tobias Hollerer.
Proceedings of the IEEE International Symposium on Mixed and Augmented Reality (ISMAR), 2015



Computing Similarity Transformations from Only Image Correspondences [code]
Chris Sweeney, Laurent Kneip, Tobias Hollerer, Matthew Turk.
Proceedings of the Conference on Computer Vision and Pattern Recognition (CVPR), 2015



Solving for Relative Pose with a Partially Known Rotation is a Quadratic Eigenvalue Problem [code]
Chris Sweeney, John Flynn, Matthew Turk.
Proceedings of the International Conference on 3D Vision (3DV), 2014 [Accepted as Oral]



gDLS: A Scalable Solution to the Generalized Pose and Scale Problem [code]
Chris Sweeney, Victor Fragoso, Tobias Hollerer, Matthew Turk.
Proceedings of the 13th European Conference on Computer Vision (ECCV), 2014

On Sampling Focal Length Values to Solve the Absolute Pose Problem [code hosted by ETH]
Torsten Sattler, Chris Sweeney, Marc Pollefeys.
Proceedings of the 13th European Conference on Computer Vision (ECCV), 2014

Model Estimation and Selection towards Unconstrained Real-Time Tracking and Mapping
Steffen Gauglitz, Chris Sweeney, Jonathan Ventura, Matthew Turk, Tobias Hollerer.
IEEE Transactions on Visualization and Computer Graphics (TVCG), vol.20, no.6, pp.825-838, June 2014

Live Tracking and Mapping from Both General and Rotation-Only Camera Motion.
Steffen Gauglitz, Chris Sweeney, Jonathan Ventura, Matthew Turk, Tobias Hollerer.
International Symposium on Mixed and Augmented Reality (ISMAR) 2012. [Best Paper Award]

HIPI: A Hadoop Image Processing Interface for Image-Based MapReduce Tasks
Chris Sweeney, Liu Liu, Sean Arietta, Jason Lawrence.
University of Virginia Undergraduate Thesis 2011. For my undergraduate thesis, I created an open-source library for the Hadoop Platform with the purpose of providing a useful research tool for large-scale image-based projects. Further details, including documentation and details on how to download HIPI can be found at this website


Software

Theia is a computer vision library for multi-view geometry specifically designed for use with SLAM and SfM. It is extremely efficient and flexible with simple interfaces for choosing and fine-tuning different descriptor, pose, matching algorithms and more. Theia provides researchers with a powerful tool to have an out-of-the-box SfM system that can easily swap in new algorithms as they are being developed and experimented. [2015 ACM Open Source Software Competition Finalist!]

Rpoly++ is a modern implementation of the popular but extremely outdated RPOLY software for computing polynomial roots. The method is an implementation of the Jenkins-Traub root-finding method for polynomials with real coefficients. It is significantly faster than alternative methods such as the Companion Matrix method while providing superior performance with respect to accuracy.

Hadoop Image Processing Interface (HIPI) is a large-scale image processing and computer vision library for Hadoop MapReduce platforms. It is specifically designed to make large-scale image tasks simple and extremely efficient. There has been wide adoption of HIPI for a variety of tasks. See the website for more information (including the mailing list).


Awards

Nobel Laureates
  • ACM Open Source Software Competition Finalist (2015)
  • NSF Graduate Research Fellowship (2013 - 2016)
  • Best Paper Award International Symposium on Mixed and Augmented Reality (ISMAR) 2012
  • Graduate Opportunity Fellowship, University of California, Santa Barbara (2011-2012)
  • Louis T. Radar Award for Outstanding Research, University of Virginia: Computer Science (2010 - 2011)
  • Google Outstanding Undergraduate Engineering Scholarship, Google (2010-2011)
  • Raven Society, University of Virginia

Other Stuff

Volunteering

I have over 2000 hours of volunteering and community service logged. I have a very close relationship with Madison House, where I am currently a member of their Alumni Council. During undergrad, I volunteered and was the leader for a trip to Treasure Beach, Jamaica in conjunction with the Treasure Beach Foundation. In the summer of 2011, I spent a month in Ghana creating clean water systems for villages with Community Water Solutions.

Comedy

Improv and stand up comedy have been hobby of mine for several years. I now perform with the Santa Barbara Improv Group have performed stand up around Santa Barbara and Los Angeles, most notably at Westside Comedy Theater, Flappers, and The Comedy Store.

Hiking

I spent a lot of time exploring the Blue Ridge mountains in Virginia, and have been working my way around the hiking trails in Santa Barbara and Los Angeles since moving out west. Outside of the US, I have been on hiking trips to the Southern Alps of New Zealand as well as the Swiss Alps.

Swimming

I have been swimming competitively since I was 5 years old, and was a member of the UVa Club Swim Team for three years. I also have a long history of coaching, having been a volunteer or assistant coach for 8 years.


© Chris Sweeney