EJShim

Machine Learning Engineer · CLO Virtual Fashion · Seoul

Eungjune Shim

I build deep-learning systems for 3D — generative models for shapes and thin shells, 3D Gaussian splatting, and geometric deep learning on meshes and scans. I like research that ships, so most of mine runs live in the browser.

fig. 1 a real-time cloth simulation — mass–spring grid, Verlet integration, computed live on your device — drag it

Research

At CLO Virtual Fashion I work on deep learning for 3D garments. My latest project, DiffGI (ECCV 2026), is a differentiable geometry-image framework that turns a single image into simulation-ready thin-shell 3D. Along the way we also made garment simulation learnable, estimating physical fabric parameters from drape tests and tag information (Eurographics 2024, CAVW 2024).

Before fashion I worked on medical AI: deep-learning pipelines for CT analysis at KIST — automated cephalometry, rotator-cuff tear classification, mandible & maxilla segmentation — then led the AI team at Imagoworks, where our tooth-segmentation and crown-generation work for 3D dental scans shipped to digital dentistry and became granted US patents.

These days I care most about 3D generative models, 3D Gaussian splatting, and edge-device inference. The browser is my favorite research medium — WebGL, WebAssembly, neural nets running on-device — like the piece of cloth hanging at the top of this page.

Selected publications

  1. 2026

    DiffGI: Differentiable Geometry Images for High-Fidelity Thin-Shell 3D Generation

    E. Shim, H. Lee, E. Ju

    ECCV 2026 arXiv project

  2. 2024

    Estimating Cloth Simulation Parameters from Tag Information and Cusick Drape Test

    E. Ju, K. Kim, S. Yoon, E. Shim, G. C. Kang, P. S. Chang, M. G. Choi

    Computer Graphics Forum 43(2) · Eurographics 2024 doi

  3. 2024

    Fast Constrained Optimization for Cloth Simulation Parameters from Static Drapes

    E. Ju, E. Shim, K. Kim, S. Yoon, M. G. Choi

    Computer Animation and Virtual Worlds 35(3) doi

  4. 2023

    Drape Simulation Estimation for Non-Linear Stiffness Model

    E. Shim, E. Ju, M. G. Choi

    Journal of the Korea Computer Graphics Society 29(3) doi

  5. 2022

    Deep Learning-Based Automatic Segmentation of Mandible and Maxilla in Multi-Center CT Images

    S. Park, H. Kim, E. Shim, B. Y. Hwang, Y. Kim, J. W. Lee, H. Seo

    Applied Sciences 12(3) doi

  6. 2020

    Web-Based Fully Automated Cephalometric Analysis by Deep Learning

    H. Kim, E. Shim, J. Park, Y.-J. Kim, U. Lee, Y. Kim

    Computer Methods and Programs in Biomedicine 194 doi 200+ citations

  7. 2020

    Automated Rotator Cuff Tear Classification Using 3D Convolutional Neural Network

    E. Shim, J. Y. Kim, J. P. Yoon, S.-Y. Ki, T. Lho, Y. Kim, S. W. Chung

    Scientific Reports 10 doi 70+ citations

  8. 2019

    Automated Maxillofacial Reconstruction Software: Development and Evaluation

    H. Kim, T. G. Son, H. Cho, E. Shim, B. Y. Hwang, J. W. Lee, Y. Kim

    Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization doi

  9. 2018

    2D–3D Registration for 3D Analysis of Lower Limb Alignment in a Weight-Bearing Condition

    E. Shim, Y. Kim, D. Lee, B. H. Lee, S. Woo, K. Lee

    Applied Mathematics — A Journal of Chinese Universities 33(1) doi

  10. 2017

    Serial Changes in 3-Dimensional Supraspinatus Muscle Volume After Rotator Cuff Repair

    S. W. Chung, K.-S. Oh, S. G. Moon, N. R. Kim, J. W. Lee, E. Shim, S. Park, Y. Kim

    The American Journal of Sports Medicine 45(10) doi

  11. 2017

    Virtual Reality and 3D Printing for Craniopagus Surgery

    G. Kim, E. Shim, H. Mohammed, Y. Kim, Y. O. Kim

    Journal of International Society for Simulation Surgery 4(1)

  12. 2016

    Virtual Surgical Planning System for Mandible Reconstruction

    H. Kim, Y. Kim, H. Cho, E. Shim, D. Lee, L. Kim, S. Park, J.-W. Lee

    Korean Journal of Computational Design and Engineering 21(2)

Full list on Google Scholar ↗

Patents

Open source & demos

EJShim/diffgi Project page for DiffGI (ECCV 2026) — image-to-3D thin-shell generation, with an on-device WebGPU demo in the works. HTML EJShim/meshless-deformations Implementation of Müller et al.’s “Meshless Deformations Based on Shape Matching” (SIGGRAPH 2005) with interactive forces. C++ ★ 12 EJShim/embedded-deformation Embedded deformation graphs (Sumner et al.) implemented three ways — Python, C++, and WASM + Svelte in the browser. Python · C++ · WASM ★ 10 EJShim/mesh_parameterization Tutte and LSCM mesh parameterization in Python, built on libigl and VTK. Python ★ 4 EJShim/tooth-segmentation-pointcloud Point-cloud deep learning (DGCNN) experiments for automated tooth segmentation of intraoral scans. Python ★ 3 EJShim/GaussianSplat_pyslang A 3D Gaussian-splatting renderer written in the Slang shading language, driven from Python via pyslang. Python · Slang

Live in your browser: 3D tooth generation · 3D face generation (CoMA) · face latent-space explorer

110 repositories on GitHub ↗  ·  research demos on YouTube ↗

Curriculum vitae

Experience

  • 2022.8 — now
    Machine Learning Engineer · CLO Virtual Fashion, Seoul

    Deep learning for 3D garments: generative thin-shell 3D (DiffGI), learnable simulation parameters.

  • 2022.5 — 2022.7
    AI Analyst · Shinhan AI, Seoul
  • 2020.6 — 2022.4
    AI Team Lead · Imagoworks, Seoul

    Deep learning for 3D dental scans — tooth segmentation & crown generation, demonstrated at IDS 2021, now US patents.

  • 2018.8 — 2020.5
    Researcher · Korea Institute of Science and Technology (KIST), Seoul

    Deep learning for medical imaging: CT segmentation, cephalometry, orthopedics.

  • 2015.1 — 2015.5
    Student Researcher · VR Lab, Konkuk University, Seoul
  • 2014.7 — 2014.8
    Intern · LG Electronics, Slough, UK

Education

  • 2016.3 — 2018.7
    M.S., Biomedical Engineering · University of Science and Technology (UST), KIST School

    Medical image analysis at KIST.

  • 2008.3 — 2016.2
    B.S., Internet & Multimedia Engineering · Konkuk University

Awards

  • 2023Best Paper Award — Korea Computer Graphics Society
  • 2018Outstanding Graduate Award — UST, KIST School
  • 2018Best Presentation Award — Korean Society of Imaging Informatics in Medicine
  • 2017Best Poster Award — Society for Computational Design and Engineering
  • 2017Best Paper Award — Asian Conference on Design and Digital Engineering
  • 2016Grand Prize (tool division), CDE Competition — Ministry of Science, ICT & Future Planning