Yuntae Jeon | 전윤태

Researcher at SKKU Advanced Institute of Nanotechnology (SAINT), South Korea

My work connects civil engineering and AI, with a focus on construction safety management and progress tracking, and is now extending to robotic applications for the field. My long-term ambition is to help build a home on Mars.

Yuntae Jeon
Scholar GitHub LinkedIn

Education

M.S., Global Smart City
Sungkyunkwan University · 2023.03–2025.02
Advisor: Prof. Seunghee Park
Korean Presidential Science Scholarship for Graduate Students
KIC Tech Frontier
UC Berkeley, Haas School of Business · 2023.07–08
B.S., Civil Engineering & Computer Science
Sungkyunkwan University · 2019.03–2023.02
Summa Cum Laude (Top 5%)
Best Graduation Paper Award (3rd Place)

Experience

Researcher — SKKU Advanced Institute of Nanotechnology (SAINT)
2026.10 – Present
Develop novel construction materials and robot control systems for use on construction sites.
AI Researcher — DeltaX
2025.02 – 2026.09
Develop, deploy, and support the operation of smart factory vision solutions for automotive press lines at Kia Motors and Hyundai Motor Company.
AI Researcher — Smartinside AI
2021.09 – 2023.02
Developed CCTV video analytics for Hyundai E&C, DL E&C, and POSCO E&C to support worker safety management, analyze heavy equipment utilization, and monitor construction progress.

Research

Civil Engineering & Architecture

  • Site representation and progress evaluation. My research combines neural radiance fields (NeRF) with building information modeling (BIM) to represent construction scenes and evaluate progress.
  • Construction safety. My work explores worker motion prediction across multiple cameras to support construction safety.

AI & Computer Vision

  • 3D scene understanding and reconstruction. My research focuses on accurate 3D object segmentation and language-driven scene interpretation using vision-language models.
  • Robust detection and tracking. My research explores low-light image enhancement for fisheye object detection, object recognition models that are robust to domain shifts, and multi-camera synchronization for cross-camera object tracking.
VLSplat: Vision-Language Guided Object-Centric 3D Gaussian Splatting via Scene Graph
Yuntae Jeon, Younho Jeon, Sujin Jin, Sungho Jo, Seunghee Park
Conference on Neural Information Processing Systems (NeurIPS) · 🎉 Spotlight (Top 1%) · 2026
Neural radiance fields for construction site scene representation and progress evaluation with BIM
Yuntae Jeon, Dai Quoc Tran, Khoa Tran Dang Vo, Jaehyun Jeon, Minsoo Park, Seunghee Park
Automation in Construction · 2025
link
Vision-Based Motion Prediction for Construction Workers Safety in Real-Time Multi-Camera System
Yuntae Jeon, Dai Quoc Tran, Almo Senja Kulinan, Taeheon Kim, Minsoo Park, Seunghee Park
Advanced Engineering Informatics · 2024
link
Leveraging Semi-Supervised Learning for Domain Adaptation: Enhancing Safety at Construction Sites through Long-Tailed Object Detection
Dai Quoc Tran, Yuntae Jeon, Armstrong Ahboah, Jinyeong Bak, Minsoo Park, Seunghee Park
ASCE's Journal of Construction Engineering and Management · 2024
link
MSD: A Benchmark Dataset for Floor Plan Generation of Building Complexes
Casper van Engelenburg, Fatemeh Mostafavi, Emanuel Kuhn, Yuntae Jeon, Michael Franzen, Matthias Standfest, Jan van Gemert, Seyran Khademi
European Conference on Computer Vision (ECCV) · 2024
Project
Low-Light Image Enhancement Framework for Improved Object Detection in Fisheye Lens Dataset
Dai Quoc Tran, Armstrong Aboah, Yuntae Jeon, Maged Shoman, Minsoo Park, Seunghee Park
IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · 2024
Paper
NeRF-Con: Neural Radiance Fields for Automated Construction Progress Monitoring
Yuntae Jeon, Almo Senja Kulinan, Dai Quoc Tran, Minsoo Park, Seunghee Park
Int. Symposium on Automation and Robotics in Construction (ISARC) · 2024
Paper
Multi-modal AI-based water pipeline data accumulation and leakage prediction research using image and ultrasound data
Yuntae Jeon, Byoungjoon Yu, Dongyoung Ko, Dai Quoc Tran, Seunghee Park
SPIE Smart Structures + Nondestructive Evaluation · 2024
Paper
Leveraging Future Trajectory Prediction for Multi-Camera People Tracking
Yuntae Jeon, Dai Quoc Tran, Minsoo Park, Seunghee Park
IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · 2023
Paper

Patents

Awards

2024.06
3rd Place — VizWiz Grand Challenge, CVPR 2024
Visual Question Answering — AnswerTherapy track
2023.10
1st Place — CVAAD Workshop, ICCV 2023
Floor Plan Auto-Completion challenge
2022.12
SKKU President's List
Selected to represent the College of Engineering
2022.12
1st Place — AI Championship
CCTV-based construction site video analysis · DL E&C project
2022.11
3rd Place — SKKU Best Graduation Paper, College of Computing
"An evolutionary neural architecture search for optimizing object detection models" — among 100+ papers
2021.11
Minister of SMEs and Startups Award (1st) — AI Championship
AI-based risk analysis for electrical facilities · KEPCO project

Scholarships

2024–25
Korean Presidential Science Scholarship for Graduate Students
1st cohort, Construction & Transportation · 제1기 대학원 대통령과학장학생 ($14,000)
2023–25
SKKU Graduate School Scholarship
2-year full scholarship ($26,000)
2022
Korean Society of Civil Engineers (KSCE) Scholarship
Park Chang-ho Scholarship ($3,000)
2022
SKKU Student Success & Creative Student Scholarship
President of SKKU ($4,000)
2021
SKKU Global Center Contributor Scholarship
SKKU Alumni Association ($2,000)
2020
Korea Institute of Registered Architects (KIRA) Scholarship
Gyeongsangbuk-do Architectural Society ($3,000)