Research Interests

    My research concern is to solve scientific problems related to Computer Graphics, Computer Animation, Computer Vision, and Medical Image.  Especially, my recent research activity is focused on human body modeling, skin/volume deformation, surface/volume reconstruction from in vivo scans (2D/3D), medical image, and general purpose GPU computation for real-time applications. Furthermore, I'm always open to any fundamental problem related to real-time rendering, photo-realistic rendering algorithms, rendering system, 3D user interfaces, and VR applications, which have been concern in previous practical projects. 

    Current Projects

    Adaptive Non-Rigid Registration of Articulated In Vivo 3D MRI in Different Poses

    • Volume registration of articulated subjects requires novel methods to accommodate the wide range of movement
      resulting from skeletal joint rotations. Articulated volume initialization based on the VSD algorithm produces good initial poses for non-rigid registration. The initialized volume is then aligned to the target volume using non-rigid volume registration with adaptively allocated sparse control points. The results are tested on in vivo 3D knee MRIs and show robust results even among volumes with significantly different poses.

    • Publications:
      • ISBI (IEEE International Symposium on Biomedical Imaging) 2008 (to appear)

    　

    Soft-tissue Deformation for In-Vivo Volume Animation

    • Volume data such as CT and MRI have obvious benefits for medical analysis and motion must be inferred from a sequence of scans. This study takes a step toward producing a continuously animatable volumetric model of a vertebrate body region using data obtained from in vivo MRI. The approach extends a recent computer graphics surface deformation technique to the volume domain.

    • Publications:
      • Proc of Pacific Graphics 2007
      • MMVR 16 (in the studies of Health Technology and Informatrics (SHTI)

    • Movie: mov[30M]

    　

    Real-Time Weighted Pose-Space Deformation on the GPU

    • WPSD (Weighted Pose Space Deformation) is an example based skinning method for articulated body animation. The per-vertex computation required in WPSD can be parallelized in a SIMD manner and implemented on a GPU. In this study, we develop a parallel deformation method using the GPU fragment processors. Joint weights for each vertex are automatically calculated from sample poses, thereby reducing manual effort and enhancing the quality of WPSD as well as SSD (Skeletal Subspace Deformation). We show sufficient speed-up of SSD, PSD and WPSD to make them suitable for real-time applications.

    • Publications : Eurographics 2006 (Graphics Forum Vol 25)

    • Movie: wmv[11M], mov[24M]

    　

    Human Hand Modeling from Surface Anatomy

    • This study presents a method for constructing a person-specific model from a single canonically posed palm image of the hand without human guidance. Tensor voting is employed to extract the principal creases on the palmar surface. Joint locations are estimated using extracted features and analysis of surface anatomy. The skin geometry of a generic 3D hand model is deformed using radial basis functions guided by correspondences to the extracted surface anatomy and hand contours. The result is a 3D model of an individual's hand, with similar joint locations, contours, and skin texture.

    • Publications: Proc. of I3D 2006

    • Movie: wmv[6.7M],  mov[11M]

     

    Previous Projects

    I performed following projects at Samsung Electronics (Research Innovation Center/E-CIM Center) as a Senior Research Engineer and Project Manager.

    Renfinity 3D: Standard Rendering S/W at Samsung Electronics (2000~2003)
    • Renfinity 3D v1.0 (around 200 copy licenses)

      Direct CAD data interface: NURBS & polygon interface with commercial CAD system

      Photo-realistic rendering algorithms: ray-tracing, a-buffer,

      Scene editor: light, camera, environment, reflection, surface attributes, 3D texture editor, and other essential functions.

      Optimized user interface for the Samsung Electronics design pipe-line
      　

    • Renfinity 3D v2.0 (around 100 copy licenses)

      Real-time Rendering: real-time shadow, real-time multi-textures display, real-time reflection/bump map, stereo display, rigid-body animation
    • VR simulation of electronic devices in the virtual interior environments ('02)
    • Distributed Rendering Library for the PC-Based Linux Cluster ('99)
    • 3D Key-frame Animation Library ('97~'98)
    • Rendering Module for Industrial 3D CAD/CAM system (VX-vision, Co-work with Varimetrix Co. in U.S.A)  (98)
      •

    VIF(Visual Interface Function) Rendering Library (1996~1999)
    • Photo-realistic rendering library (Co-worked with Visual Innovation Co. in U.S.A.)

      　

    • Library interface similar to Renderman specifications:
      • Object oriented C/C++ Interface for windows and UNIX system
      • Raytracing, A-buffer, radiosity rendering
      • Geometry interfaces for surface(NURBS) and polygon
      • Texture mapping & procedural textures
      • Script based scene editing
        　

    Before Joining Samsung Electronics, I did following projects at Computer Graphics Lab in Sogang University as a research assistant of Prof. In-Sung Ihm during my MS program.

    Distributed Volume Rendering  (1994~1996)
    • Visualization of 3D Medical Data Using Distributed Volume Rendering Technique on the Information Super-Highway. 

      　

    • Design and Implementation of Volume Data Rendering Toolkit.

      　

    • Volume Visualization of Visible Human Data Set
      •