Jin-Xiang Chai

Jinxiang Chai (柴金祥)

Associate Professor

Computer Science and Engineering

Texas A&M University

College Station,

Texas 77843-3112, USA

Office: 527D H.R. Bright

Phone: 979-845-3510

Fax: 979-845-1420

Mocap Lab: 505 H.R. Bright

Email: jchai@cs.tamu.edu

Symposium on Computer Animation:

I am helping to organize ACM SIGGRAPGH / Eurographics Symposium on Computer Animation 2013, July 19-21 in Anaheim, just before SIGGRAPH. The paper submission deadline is next Tuesday (April 16)!

Please submit your best work to sca2013.cs.tamu.edu/.

SHORT BIO:

Jinxiang Chai is currently an associate professor in the Department of Computer Science and Engineering at Texas A&M University. He received his Ph.D in 2006 from the Robotics Institute at Carnegie Mellon. His primary research is in the area of computer graphics and vision with broad applications in other disciplines such as robotics, human computer interaction, and biomechanics, virtual and augmented reality. He is particularly interested in developing representations and efficient computational models that allow acquisition, analysis, understanding, simulation, and control of natural human motion, including movements of the body, face and hand, multi-person interaction, and crowds. He draws on ideas from graphics, vision, machine learning, robotics, biomechanics, neuroscience and applied math. He received an NSF CAREER award for his work on theory and practice of Bayesian motion synthesis.

RESEARCH PROJECTS AND INTERESTS:

Human motion modeling, understanding, simulation and control. The primary goals of our research herein are: (1) to develop efficient representations and computational models to analyze how humans move by utilizing prerecorded motion data, physics, biomechanics principles, and control theories; and (2) to apply the new models to solve important and challenging problems in computer graphics, computer vision and robotics such as synthesis and control of animated human characters, human motion tracking and classification, and motion planning and control for humanoid robots.

Motion capture for everyone. We are developing next-generation motion capture technologies that minimize the cost and intrusiveness of motion capture so that the technology is practical and easily accessible to every home user. Notable examples include full-body performance capture using low-cost sensors, video-based motion capture, and realtime motion capture using RGBD cameras. We are also interested in developing new techniques for capturing the physics of complex movements from real world.

Facial acquisition, modeling and animation. One of holy grail problems in computer graphics has been the realistic animation of the human face. We have been developing new methods to animate and control virtual faces by capturing and analyzing facial performances of real people.

Hand motion acquisition, modeling and simulation. We are exploring new techniques for capturing high-quality hand gestures, grasping and manipulation for hand motion understanding, modeling, simulation and control.

Visual modeling and understanding. We seek to build systems that can visually model and understand complex movements such as full-body movements, facial deformations, hand gestures, multi-person interaction, or animal movements. For example, we have developed new techniques for modeling deformable objects and articulated bodies using a single camera. Before that, I had been working on image-based modeling and rendering.

Data-driven graphics and vision. I am genuinely interested in data-driven approaches for solving ill-posed graphics and vision problems, such as animation control, deformation modeling, and vision-based motion tracking. I am particularly interested in learning techniques that can scale up to massive and heterogeneous datasets.

TEACHING:

CSCE 441 (Spring 2013): Computer Graphics

CSCE 643 (Spring 2012): Computer Vision

SELECTED RECENT PUBLICATIONS: (For more details, see my Projects page)

Video-based Hand Manipulation Capture Through Composite Motion Control

ACM Transactions on Graphics (to present at SIGGRAPH 2013)

Generative Statistical Models for Semantic Motion Analysis and Synthesis

ACM Transactions on Graphics (Presented at SIGGRAPH ASIA 2012)

Accurate Realtime Full-body Motion Capture Using a Single Depth Camera

ACM Transactions on Graphics (Presented at SIGGRAPH ASIA 2012)

Combining Marker-based Mocap and RGB-D Camera for Acquiring High-fidelity Hand Motion Data

ACM SIGGRAPH/Eurographics Symposium on Computer Animation (SCA 2012)