Xiaojing Chen

Ph.D Candidate in Computer Science

Center for Research in Intelligent Systems

University of California, Riverside, CA 92521

Email: xchen010-AT-cs.ucr.edu

[Research | Publications | Internships | Awards | Services and Activities | TAs]


Short Biography:

I am currently a graduate student researcher advised by Prof. Bir Bhanu at the Center for Research in Intelligent Systems, University of California, Riverside. Before I came to US, I spent some time in Europe and received MSc in Computer Science (with Honors) from Leiden University in the Netherlands in 2009. I did my undergraduate study in Information Management and Systems at Beijing Language and Culture University in China.

My research interests include machine learning, computer vision, video analysis, and pattern recognition. My recent research involves large-scale face image retrieval, automatic multi-target tracking for both single camera and multiple cameras.


Research 

  • Soft-biometrics Integrated Mutli-target Tracking [C-06]

    We present a soft biometrics based appearance model for multi-target tracking in a single camera. We use soft biometrics which are invariant to view and illumination changes to learn a discriminative appearance model in an online manner. Compared to low level features, soft biometrics are robust against appearance variation.

  • Elementary Grouping Model for Multi-target Tracking [C-05]

    Multi-target tracking is to recover trajectories of all targets while maintaining identity labels consistent. Beside low level infomation, we learn elementary groups, i.e. groups that contain only two targets, for inferring high level context to improve multi-target tracking. The proposed method is efficient, handles group merge and split, and can be easily integrated into any basic affinity model.

  • Reference Set Based Appearance Model for Tracking in Multiple Cameras [C-04] [J-01]

    Multi-target tracking in non-overlapping cameras is challenging due to the vast appearance change of the targets across camera views. Therefore, direct track association is difficult and prone to error. We propose a novel reference set based appearance model to improve multi-target tracking in a network of nonoverlapping cameras.

  • Multi-level Features for BoW Based Large-scale Face Image Retrieval [C-02]

    To improve the retrieval accuracy of the face images in large-scale dataset, we introduce a novel multi-level feature extraction method to bag-of-words (BoW) based retrieval system. It employs various scales of features simultaneously to encode different texture information and emphasizes image patches that are more discriminative as parts of the face.


    Publications 

    Book Chapter

      [B-01]    Xiaojing Chen, Le An, Bir Bhanu, "Soft-biometrics and Reference Set Integrated Model for Tracking Across Cameras", Distributed Embedded Smart Cameras, Springer, 2014. (PDF)

    Journal Articles

      [J-01]    Xiaojing Chen, Le An, Bir Bhanu, "Multi-Target Tracking in Non-overlapping Cameras Using a Reference Set", IEEE Sensors Journal 2014, Article DOI: 10.1109/JSEN.2015.2392781, (PDF)

    Conference Proceedings

      [C-06]    Xiaojing Chen, Bir Bhanu, "Soft Biometrics Integrated Multi-target Tracking", IEEE International Conference on Pattern Recognition (ICPR), 2014. (PDF)

      [C-05]    Xiaojing Chen, Zhen Qin, Le An, Bir Bhanu, "An Online Learned Elementary Grouping Model for Multi-target Tracking", IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2014. (PDF)

      [C-04]    Xiaojing Chen, Le An, Bir Bhanu, "Reference Set Based Appearance Model for Tracking Across Non-overlapping Cameras", ACM/IEEE International Conference on Distributed Smart Cameras (ICDSC), 2013. (Oral) (PDF)

      [C-03]    Le An, Xiaojing Chen, Mehran Kafai, Songfan Yang, Bir Bhanu, "Improving Person Re-Identification by Soft Biometrics Based Reranking", ACM/IEEE International Conference on Distributed Smart Cameras (ICDSC), 2013. (Oral) (PDF)

      [C-02]    Xiaojing Chen, Le An, Bir Bhanu, "Improving Large-Scale Face Image Retrieval Using Multi-Level Features", IEEE International Conference on Image Processing (ICIP), 2013. (PDF)

      [C-01]    Devrim Unay, Xiaojing Chen, Aytul Ercil, Mujdat Cetin, Radu Jasinschi, Mark A. van Buchem, Ahmet Ekin, "Binary and nonbinary description of hypointensity for search and retrieval of brain MR images", IS&T/SPIE Electronic Imaging, Multimedia Content Access, 2009. (Oral) (PDF)


    Internships 

    • Research Intern, Philips Research, Eindhoven, The Netherlands, January 2009 - July 2009

      Proposed an algorithm to extract different parts of basal ganglia region from MRI scans.

      Estimated hypointensity load in normal adults using MRI images.


    • Research Intern, Philips Research, Eindhoven, The Netherlands, June 2008 - December 2008

      Proposed an improved binary hypo-intensity and a novel non-binary hypo-intensity description.

      Further tested on Golden Datasets substantiated the robustness and reliability of both methods.



    Awards 

    • Graduate Division Fellowship Award, University of California at Riverside, 2010

    • Graduation with cum laude, Leiden University, 2009

    • First-class Comprehensive Scholarship, Beijing Language and Culture University, 2006


    Services and Activities 

    • Reviewer for IEEE Transactions on Image Processing

    • Reviewer for IEEE Sensors Journal

    • Reviewer for Elsevier Journal of Computer Vision and Image Understanding

    • Reviewer for Elsevier Journal of Pattern Recognition

    • Reviewer for Elsevier Journal of Information Sciences

    • Reviewer for SPIE Journal of Electronic Imaging

    • Reviewer for the IEEE Conferences: ICIP, ICPR, CVPR, ICDSC, WACV


    TAs 

    • CS 008: Introduction to Computing

    • CS 161: Design and Analysis of Computer Architecture

    • CS 181: Principles of Programming Languages

    [Research | Publications | Internships | Awards | Services and Activities | TAs]