cv

CV as of Jul 23, 2026.

Table of contents

Basics

Name Taeho Kang
Label Ph.D. Candidate
Email tho.kang.7@gmail.com
Url https://tho-kn.github.io
Summary I develop computer vision and graphics methods for interactive 3D and XR. My work covers egocentric full-body pose estimation from headset cameras, dynamic 3D scene reconstruction with Gaussian Splatting, and pose-driven animation of 3D scenes. My goal is interactive 3D content that responds to a user's body without specialized capture hardware.

Education

  • 2021.03 - Present

    Seoul, Korea
    Ph.D. Candidate
    Seoul National University
    Computer Science and Engineering
    • Advisor: Youngki Lee
  • 2017.03 - 2021.02

    Seoul, Korea
    Undergraduate Student
    Seoul National University
    Computer Science and Engineering
    • Cum Laude

Publications

Research

  • 2022.02 - Present
    Research Student
    Seoul National University · Seoul, Korea
    Human-Centered Computer Systems Lab
    Conducting research on 3D computer vision for human scene understanding.
  • 2021.03 - 2022.01
    Research Student
    Seoul National University · Seoul, Korea
    Movement Research Lab
    Conducted research on computer graphics, particularly human motion.
  • 2019.07 - 2019.08
    UROP Research Intern
    Seoul National University · Seoul, Korea
    Movement Research Lab
    Worked on a soft-body simulation model based on mass-spring systems for physically-based animation.

Work

  • 2024.07 - 2024.11
    Applied Scientist Intern
    Amazon.com Services LLC · United States
    Conducted research on computer vision models for navigation assistance.
  • 2020.01 - 2020.12
    Contract Software Developer
    SAMATTI · Korea
    Developed a custom garment annotation tool using PyQt to streamline data labeling processes.
  • 2017.12 - 2018.02
    Intern
    Hutom · Korea
    Participated in a project that involved organ simulations and deep learning-based for surgical assistance systems.

Volunteer

  • 2024.01 - Present
    Reviewer
    Academic Conferences
    SIGGRAPH Asia 2024
    CVPR 2025
    ICCV 2025
    IMWUT 2025 February
    NeurIPS 2025
    ICLR 2026

Skills

Research Interests
Egocentric Full-Body Pose Estimation
Pose-Driven 3D Scene Animation
3D Scene Reconstruction & Generation
Technical Skills
Deep Learning & Vision: PyTorch, Python, C++, CUDA, 3D Geometry
Graphics & Simulation: Unreal Engine, OpenGL, Gaussian Splatting
Research Engineering: Motion Capture Pipeline Automation, Custom Annotation Tool Development (PyQt)