Xuemei Zhao                                                   

Welcome to my homepage.


Research Interest

Computer Vision, Machine Learning


2008.8-present            University of Souther California              Los Angeles, CA  

                                    PhD in Electrical Engineering

                                    Advisor:  Prof. Gerard Medioni

2004.8-2008.7            Tsinghua University                                 Beijing, China

                                    Bachelor in Electronic Engineering


2012.5-2012.8            Google, Image Search Team                 Mountain View, CA

Selected Publications

1. Xuemei Zhao, Dian Gong and Gerard Medioni, "Tracking Using Motion Patterns for Very Crowded Scenes", Proc. of the 12th European Conference on Computer Vision (ECCV 2012), Firenze, Italy, October 2012.

2. Dian Gong, Gerard Medioni, Sikai Zhu and Xuemei Zhao, "Kernelized Temporal Cut for Online Tempoal Segmentation and Recognition", Proc. of the 12th European Conference on Computer Vision (ECCV 2012), Firenze, Italy, October 2012.

3. Dian Gong, Xuemei Zhao and Gerard Medioni, "Robust Multiple Manifolds Structure Learning", Proc. of the 29th International Conference on Machine Learning (ICML 2012), Edinburgh, Scotland, June 2012.

4. Jan Prokaj, Xuemei Zhao and Gerard Medioni, "Tracking Many Vehicles in Wide Area Aerial Surveillance", Workshop on Camera Networks and Wide Area Scene Analysis joint with IEEE CVPR 2012, Rhode Island, USA, June 2012.

5. Xuemei Zhao and Gerard Medioni, "Robust Unsupervised Motion Pattern Inference from Video and Applications", Proc. of the IEEE 13th International Conference on Computer Vision (ICCV 2011), Barcelona, Spain, November 2011.  

6. Dian Gong, Xuemei Zhao and Qiong Yang, "Sparse Non-Negative Pattern Learning for Image Represenation", Proc. of the 15th IEEE International Conference on Image Processing (ICIP 2008) ,San Diego, California, USA, October 2008. 


This software library (download) provides both CPU and GPU implementation of Tensor Voting's ball vote and stick vote. In the sample folder, sample code, sample input and output files are provided. Since this is a preliminary release, please let us know if you find any bug. Thank you.

Contact Me

Xuemei zhao
Email: xuemeiz at usc dot edu

Office: 3737 Watt Way, USC PHE 101, Los Angeles, CA, 90089


The University of Southern California does not screen or control the content on this website and thus does not guarantee the accuracy, integrity, or quality of such content. All content on this website is provided by and is the sole responsibility of the person from which such content originated, and such content does not necessarily reflect the opinions of the University administration or the Board of Trustees