Jingjing Liu (Danny)
Department of Computer Science, Rutgers University,
110 Frelinghuysen Road, Piscataway, NJ 08854-8019
Email: jl1322 at cs dot rutgers dot edu
[Home] [Education] [Working Experience] [Projects] [Publications] [TA Work] [Courses] [Others]

Jingjing Liu is a PhD student in Computer Science at Rutgers, the State University of New Jersey. He is a research assistant working with his academic advisor Prof. Dimitris N. Metaxas in Computational Biomedicine Imaging and Modeling Center (CBIM).

His research interests are computer vision and machine learning, specially on facial expression analysis and pedestrian detection.

(Curriculum Vitae).

  • 09/2011 ~ Present, PhD student in Computer Science, Rutgers University
  • 09/2008 ~ 07/2011, M.S., Computer Science and Technology, Chinese Academy of Sciences, Beijing, China
  • 09/2003 ~ 07/2007, B.E., Automatic Control and Information Technology, Beihang University, Beijing, China

  • Working Experience
  • 07/2012 ~ present, Research Assistant at Rutgers University
  • 06/2015 ~ 08/2015, Software Engineer Intern at Apple Inc., Cupertino, CA
  • 05/2014 ~ 12/2014, Research Intern at IBM T.J. Watson Research Center, Yorktown Heights, NY
  • 09/2011 ~ 05/2012, Teaching Assistant at Rutgers University
  • 09/2009 ~ 07/2011, Research Assistant at NLPR, Chinese Academy of Sciences
  • Jingjing Liu (刘京京)
    Department of Computer Science Rutgers University
    Piscataway, NJ 08854

    Robust Pedestrian Detection

    People detection in images is a fundamental vision problem, which is central to a wide range of applications such as video surveillance, robotics, and autonomous driving. However, many challenges, including occlusion, low image resolution, and cluttered background, prevent artificial vision systems from approaching human-level perception ability on identifying pedestrians. In these work, we are targeting robust pedestrian detection algorithms for around-the-clock applications, by leveraging contextual information and multispectral imaging.


    Scalable Mammogram Retrieval

    Mammogram analysis is known to provide early-stage diagnosis of breast cancer by reducing its morbidity and mortality. We aim to design a scalable content-based image retrieval (CBIR) system for the analysis of digital mammogram. CBIR is of great significance for breast cancer diagnosis because it provides doctors image-guided avenues to access relevant cases. Diagnosis and clinical decisions based on such cases offer reliable and consistent references for doctors. Our system is readily applicable to large-scale mammogram databases so that a high number of analogical cases would be returned as a clinical supplement.


    Non-maunal events in American Sign Language

    Changes in eyebrow configuration, in conjunction with other facial expressions and head gestures, are used to signal essential grammatical information in signed languages. We propose an automatic recognition system for non-manual grammatical markers in American Sign Language (ASL) based on a multi-scale, spatio-temporal analysis of head pose and facial expressions.

    Global sea level prediction

    This is course project of Pattern Recognition working on global sea level prediction. In this project, I used Gaussian Process Regression (GPR) model with spatial-temporal covariance. We conducted experiments on a public sea level dataset, including global sea levels of more than 20 years. The GPR model can precisely predict the trend of sea level changes. The most interesting thing is that our predictive model correctly discover the El Niño event in 1997-’98, the strongest El Niño events occurred in the 20th century.


    Crowd counting

    Pedestrian counting has been a challenging topic especially in video surveillance for a long time due to the view variations, scale changes and spatial occlusions. We formulate the problem of pedestrian counting as a joint maximum a posteriori (MAP) problem. Markov Chain Monte Carlo (MCMC) is utilized to search for an optimal configuration set to match the spatio-temporal context.

    Pedestrian couting with ceiling camera

    In this project, we used the directional chamfer matching method for pedestrian counting, based on ceiling cameras. Circles of multiple sizes are used as the shape models of pedestrians, since seldom occlusions happen in such perspective. We tracked individuals and the trajectories in the ROI are counted as the number of pedestrians.

    TA Work
    No Image

    CS214, System Programming

    TA for this course in Spring 2016. The instructor is Prof. John-Austen Francisco.

    • The aim of this course is to introduce the student to the process of writing low-level programs that interact directly with a computer's operating system and hardware, as well as to develop the student's ability to build large applications in a team environment. Upon completion of this course, the successful student should be able to design, write, test, and analyze moderately complicated programs using the C programming language and UNIX/Linux operating systems.

    No Image

    CS21, Computer Architecture

    TA for this course in Spring 2015. The instructor is Prof. Santosh Nagarakatte. Contents include:

    • C programming; Assembly language techniques; instruction-set design; Boolean algebra; digital logic; The memory hierarchy; Input/Output concepts.

    No Image

    CS111, Introduction to Computer Science

    TA for this course in Fall 2012 and Spring 2012. The instructor is Prof. Sesh Venugopal and Dr. Andrew Tjang. Contents include:

    • Fundamentals of Programming; Introductory Java Programming; Procedural Programming; Simple Data Structures; Efficient Algorithms.