Hwan Heo

Lead 3D AI Research Engineer

NC AI · VARCO 3D

Building production-scale 3D generative systems across native representations, large-model training, scalable inference, and graphics algorithms.

Projects Blog About

Projects

Production 3D AI systems · 4 papers · 6 invited talks

Project Index

Nine VARCO 3D community assets alternating between textured and geometry views
Production R&D VARCO 3D NC AI’s proprietary 3D generative AI service, delivering detailed 1024³ mesh generation, 4K texture synthesis, and production-ready low-poly assets.
Contribution
Trained and accelerated geometry models for production; built CUDA kernels for remeshing, UV unwrapping, and rasterization-based multi-view-to-texture baking.
3D GenerationCUDA NC AI / 2025–Present 30K+ users · 300K+ assets generated
CaPa geometry, Janus-free multi-view painting, and final textured mesh pipeline
Research CaPa Designed Spatially Decoupled Attention to solve the Janus problem, enabling the first 4K-texture image-to-3D system and leading to VARCO 3D. DiffusionTextured Mesh NCSOFT / 2024 Featured on Hugging Face Daily Papers
Hash-grid gradient and trilinear interpolation diagrams
Research Instant Pose for Hash-Grid NeRFs Identified unstable hash-grid coordinate gradients and designed a smooth backward interpolation path for fast camera-pose refinement, published at ICML 2023. NeRFCamera Pose NAVER AI Lab / 2022–2023 ICML 2023
Research Selected Papers 4 publications

  1. CaPa: Carve-n-Paint Synthesis for Efficient 4K Textured Mesh Generation

    Hwan Heo, Jangyeong Kim, Seongyeong Lee, Jeong-A Wi, Junyoung Choi, Sangjun Ahn

    Technical Report, Featured on Hugging Face Daily Papers, TLDR AI

    paper project github

  2. DreamSurfels: Generative Gaussian Surfels for Geometrically Accurate 3D Content Creation

    Hwan Heo, Sangjun Ahn

    The 15th International Conference on ICT Convergence (ICTC), Oct 2024
    Oral presentation, domestic conference

  3. Robust Camera Pose Refinement for Multi-Resolution Hash Encoding

    Hwan Heo, Taekyung Kim, J. Lee, Jiyoung Lee, Soohyun Kim, Hyunwoo J. Kim, Jin-Hwa Kim

    The 40th International Conference on Machine Learning (ICML), Jul 2023

    paper project

  4. Semantic-aware Occlusion Filtering Neural Radiance Fields in the Wild

    Jaewon Lee, Injae Kim, Hwan Heo, Hyunwoo J. Kim

    preprint, arXiv 2023

    paper project github
Speaking Invited Talks 6 talks

  • Introduction to 3D Generative AI

    Generative AI Seminar (CS.93000B) - KAIST

    PDF

  • Introduction to 3D Generative AI

    Machine Learning for Computer Vision - KAIST

  • Spatial 3D AI Technology and its Applications

    K-GEO Festa - Ministry of Land, Infrastructure and Transport in Korea (국토교통부)

  • Radiance Fields and Spatial 3D AI for Digital Twin

    National Geographic Information Institute in Korea (국토지리정보원)

  • Rasterization vs. Ray Tracing with 3D Gaussian Primitives

    NCSOFT

  • Joint Optimization of Camera Pose and Accelerated Radiance Fields

    NAVER Cloud

Technical Blog

Essays and technical notes on building 3D generative systems:
from representations and model training to CUDA kernels, geometry processing, and production engineering.

View all posts

Selected Writing

About

Hello, I’m Hwan Heo. I work where generative models, systems, and graphics meet.

Portrait of Hwan Heo

Hwan Heo

Lead 3D AI Research Engineer NC AI · VARCO 3D

At NC AI, I lead research and engineering for the VARCO 3D, developing production 3D AI systems across large-scale generative model training, inference optimization, and graphics algorithms.

When I'm away from CUDA kernels and mesh pipelines, I'm usually playing guitar, making music, or following an unexpected STEM rabbit hole.

Document Curriculum Vitae Download
Career Experience & Education View details Close

Experience

Lead 3D AI Research Engineer

3D Generation, NC AI

  • Leading VARCO 3D R&D across generative model training, inference optimization, graphics algorithms, and deployment.

AI Researcher

Neural Rendering & 3D Generation, NCSOFT Research
Research Personnel · Alternative Military Service

  • Developed 3D generation and neural rendering systems spanning latent diffusion, Gaussian Splatting, and NeRF avatars.

Research Intern

Generation Research, NAVER AI Lab

  • Researched pose optimization for neural rendering.
  • Mentored by Dr. Jin-Hwa Kim.

Education

M.S. in Computer Science & Engineering

MLV Lab, Korea University, Seoul

B.S. in Computer Science & Engineering

Korea University, Seoul

Hwan Heo · 3D AI