Jie Sun

 

College of Computing

 

Georgia Institute of Technology

 

 

Biography

I am a PhD candidate in College of Computing, Georgia Institute of Technology. My research focuses on applying machine learning to computer vision and robot perception. My PhD thesis is about affordance perception and categorization. My advisors are James M. Rehg and Aaron Bobick.

Current Research

General object categorization and affordance perception

Past Projects

Traversability prediction and “learning from examples” for ground robot navigation (in the LAGR project)

Terrain synthesis from example elevation data

Frontal face detection with cascade classifiers

Prediction of classification performance (cumulative match characteristic) on large population data from small population data

Publications

1.     S. C. Brubaker, J. Wu, J. Sun, M. D. Mullin, and J. M. Rehg, On the Design of Cascades of Boosted Ensembles for Face Detection, Int. J. of Computer Vision, Special Issue on Learning for Vision 2007

2.      Howard Zhou, Jie Sun, Greg Turk, and James Rehg, Terrain Synthesis from Digital Elevation Models, IEEE Trans. Visualization and Computer Graphics (TVCG) 2007 (PDF)

3.      Jie Sun, Tejas Mehta, Dave Wooden, Matt Powers, James Regh, Tucker Balch, and Magnus Egerstedt, Learning from Examples in Unstructured, Outdoor Environments, J. of Field Robotics (JFR) 2006 (PDF)

4.      Dongshin Kim, Jie Sun, Sangmin Oh, James Rehg, Aaron Bobick, Traversability Classification Using Unsupervised On-Line Visual Learning for Outdoor Robot Navigation, IEEE ICRA 2006 (PDF)

5.      Jie Sun, James Rehg, Aaron Bobick, Automatic Cascade Training with Perturbation Bias,  IEEE CVPR 2004 (PDF) (Oral Presentation Talk slides)

6.      Amos Johnson, Jie Sun, Aaron Bobick, Using Similarity Scores from a Small Gallery to Estimate Recognition Performance for Larger Galleries, IEEE Int. Workshop on Analysis and Modeling of Faces and Gestures, 2003. (PDF)

7.      Amos Johnson, Jie Sun, Aaron Bobick, Predicting Large Population Data Cumulative Match Characteristic Performance from Small Population Data, the 4th International Conference on Audio and Video Based Biometric Person Authentication (AVBPA) 2003 (PDF)

Contact Information
Office: 230 Tech Square Research Building
Phone: 404 385 4228

Email: sun at cc.gatech.edu