Seonwook (Wookie) Park

my_pic.jpg

I am a Senior Research Scientist at NVIDIA Research, where I am part of the AI-Mediated Reality and Interaction (AMRI) research group. I am particularly interested in computational human perception with a focus in measuring and understanding human gaze.

I completed my PhD in the Advanced Interactive Technologies lab at ETH Zurich, advised by Prof. Dr. Otmar Hilliges. As part of my research, I proposed novel computer vision and machine learning techniques as well as image and video datasets for the task of webcam-based gaze estimation. Prior to joining NVIDIA, I worked at Lunit as a Research Scientist where I worked on topics in microscope image analysis and computational pathology (CPATH).

You can reach me via seon.wook@{this_domain} and my latest CV can be found here.

News

Jan 14, 2026 We are hosting two workshops at CVPR 2026: one workshop on Interactive Physical AI (IPA 2026) and one workshop on Eye and Gaze in Computer Vision (GAZE 2026) - both will receive full paper submissions.
Jun 24, 2024 I joined NVIDIA Research as a Senior Research Scientist. I will be working on diverse topics in Computer Vision and Deep Learning with a focus on human perception and understanding.

Recent Publications

  1. 2026_arxiv_videofdb.jpg
    VideoFDB: Evaluating Full-Duplex Vision-Speech Capabilities in Conversational Agents
    Amrita Mazumdar, Seonwook Park, Rajarshi Roy, Nikhil Srihari, Shengze Wang, Yuhao Zhou, and 3 more authors
    arXiv preprint arXiv:2605.30256, 2026
  2. 2026_neurips_dyaplex.jpg
    DyaPlex: Full-Duplex Speech-Motion Model for Dyadic Interaction
    Koki Nagano*, Hongyu Liu*, Seonwook Park, Tianye Li, Amrita Mazumdar, Christian Jacobsen, and 6 more authors
    In NeurIPS , 2026
  3. 2026_neurips_exo2ego.jpg
    Grounded-Exo2Ego: Structured Semantic Grounding for Robust Exocentric-to-Egocentric Video Generation
    Shengze Wang, Michael Stengel, Tianye Li, Seonwook Park, Amrita Mazumdar, Koki Nagano, and 2 more authors
    In NeurIPS , 2026
  4. DETRAM: End-to-end DEtection, Tracking and Recovery of HumAn Meshes
    Chunggi Lee, Seonwook Park , Wanhua Li, Umar Iqbal, and Hanspeter Pfister
    In ECCV , 2026
  5. Instant Expressive Gaussian Head Avatars at Over 100 FPS
    In ECCV , 2026
  6. 2025_siggraph_gaia.jpg
    GAIA: Generative Animatable Interactive Avatars with Expression-conditioned Gaussians
    Zhengming Yu, Tianye Li, Jingxiang Sun, Omer Shapira, Seonwook Park, Michael Stengel, and 5 more authors
    In SIGGRAPH Conference Papers , 2025