Search this site
Embedded Files
Sungho Chun

Sungho Chun (전성호)

Ph.D. Candidate

Computer Vision Lab.

Kwangwoon University, Seoul, Korea


e-mail : asw9161@gmail.com

[Google Scholar] [LinkedIn] [CV] [Github]

NEWS

  • September 2026: A paper was accepted to NeurIPS 2026 🎉

  • March 2026: A paper was accepted to Pattern Recognition (PR) 🎉

  • August 2025: I was awarded a doctoral research grant from the National Research Foundation of Korea (NRF) 🎉

  • August 2025: I will be presenting TRG at KCCV 2025 in Busan. Feel free to chat with me! 😊

  • December 2024: I will join HCVL at the University of Birmingham in the UK as a Visiting Researcher 😊

  • July 2024: A paper was accepted to ECCV 2024 🎉

  • June 2023: A paper was accepted to ICIP 2023 🎉

  • August 2022: A paper was accepted to WACV 2023 🎉

RESEARCH INTERESTS

Computer Vision, Machine Learning, Deep Learning

  • 3D Vision

    • Multi-view geometry

    • Perspective geometry

  • Human pose and shape estimation

    • 6DoF head pose estimation

    • 3D face reconstruction

    • Occlusion-robust pose estimation

    • Detection

  • Dataset curation & annotation

PUBLICATIONS

HOGWARTS: Mitigating Perspective Distortion for 6DoF Head Pose Estimation via Virtual Camera Space

Sungho Chun, Ju Yong Chang

Advances in Neural Information Processing Systems (NeurIPS), 2026

[Paper][Code]


Bidirectional Regression for Monocular 6DoF Head Pose Estimation and Reference System Alignment

Sungho Chun, Boeun Kim, Hyung Jin Chang, Ju Yong Chang

Pattern Recognition (PR, IF: 9.1), vol. 179, Part A, Art. no. 113585, 2026

[Paper][Code]


6DoF Head Pose Estimation through Explicit Bidirectional Interaction with Face Geometry

Sungho Chun and Ju Yong Chang

European Conference on Computer Vision (ECCV), 2024

[Paper][Video][Demo][Code]


Representation learning of vertex heatmaps for 3d human mesh reconstruction from multi-view images

Sungho Chun, Sungbum Park, and Ju Yong Chang

IEEE International Conference on Image Processing (ICIP), 2023

[Paper][Video]

Learnable human mesh triangulation for 3D human pose and shape estimation

Sungho Chun, Sungbum Park, and Ju Yong Chang

IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2023

[Paper] [Video][Project Page]

EXPERIENCE

Visiting Researcher Dec.2024 - Mar.2025

  • Visiting researcher at University of Birmingham, UK (Host: Prof. Hyung Jin Chang)

Robot development Internship Jan.2020 - Feb.2020

  • R&D team at Robomation, Seoul, Korea

  • Product development and testing

Academic club Mar.2018 - Feb.2021

  • BARAM, Kwangwoon University

  • Built and demonstrated robots each year

  • Led a robotics team and won an award at a national competition

Military Service Feb.2016 - Nov.2017

  • Republic of Korea Army, Honorable Discharged 

Projects

Research Project Apr.2026 - Current

  • KEIT, Daegu, Korea

  • Topic : Development of On-Device AI-Based Multi-Sensor Fusion Intelligent Perception Technology for Ensuring Worker Safety in Manufacturing Logistics Processes

  • Role: Develop and implement a deep learning algorithm


Research Project Mar.2026 - Current

  • ETRI, Daejeon, Korea

  • Topic : Full-Range Head Pose Estimation and Gaze Following in Crowd Scenes

  • Role: Develop and implement a deep learning algorithm

  • Head detection and pose estimation in crowded scenes

  • Skills: Dataset Curation & Annotation, Occlusion-robust pose estimation, Pose estimation from low-resolution image, Object detection 


Research Project Jun.2025 - Nov.2025

  • ETRI, Daejeon, Korea

  • Topic : Development of Long-Range Head Object Analysis Technology for Crowd Analysis

  • Role: Develop and implement a deep learning algorithm

  • Head detection and pose estimation in crowded scenes

  • Skills: Occlusion-robust pose estimation, Pose estimation from low-resolution image, Object detection 


Research Project Apr.2023 - Dec.2025

  • ETRI, Daejeon, Korea

  • Topic : FACS-compatible digital human facial expression style transfer technology

  • Role: Develop and implement a deep learning algorithm

  • Reconstructing 3D face mesh and estimating 6DoF head pose from monocular image

  • Achieving state-of-the-art in 6DoF head pose estimation task

  • 1 paper accepted to ECCV 2024

  • 1 paper accepted to PR 2026

  • Skills: Perspective geometry, 6DoF head pose estimation, 3D face mesh reconstruction 


Research Project Apr.2022 - Oct.2022

  • NCSOFT, Seoul, Korea

  • Topic : 3D human mesh reconstruction from multi-view video for animation [Video]

  • Role: Develop and implement a deep learning algorithm

  • Occlusion-robust and temporally consistent human pose estimation from multi-view video

  • Achieving state-of-the-art in human pose and shape estimation task

  • 1 paper accepted to ICIP2023

  • Skills: Multi-view geometry, Triangulation, Representation learning, Human pose and shape estimation, Temporal refinement


Research Project Aug.2021 - Jan.2022

  • NCSOFT, Seoul, Korea

  • Topic : Physically and temporally consistent human pose estimation from multi-view video

  • Role: Develop and implement a deep learning algorithm

  • Human pose and shape estimation from multi-view images

  • Achieving state-of-the-art in terms of predicting joint rotation and human body shape

  • 1 paper accepted to WACV2023, 1 domestic patent applied

  • Skills: Multi-view geometry, Triangulation, Human pose and shape estimation

EDUCATION

Ph.D., Electronics and Communications Engineering Mar.2023 - 

  • Kwangwoon University, Seoul, Korea

  • Advisor: Prof. Ju Yong Chang

M.S., Electronics and Communications Engineering Mar.2021 - Feb.2023

  • Kwangwoon University, Seoul, Korea

  • Advisor: Prof. Ju Yong Chang

  • GPA score : 4.42/4.5 

B.S., Robotics Mar.2015 - Feb.2021

  • Kwangwoon University, Seoul, Korea

  • GPA score : 4.06/4.5 

PROFESSIONAL SERVICES

Reviewer

  • IEEE Transactions on Image Processing (TIP)

PATENT

  • Young Hee Kim, Ju Yong Chang, and Sung Ho Chun, “Method and Apparatus for Providing 6DoF Head Pose Estimation from Monocular Video,” U.S. Patent Application No. 19/254,563, published May 14, 2026.

  • SangJun Ahn, Sunwon Jeong, Ju Yong Chang, and Sungho Chun, “Electronic Device, Method, and Computer-Readable Storage Medium for Acquiring Information Indicating Shape of Body from One or More Images,” U.S. Patent Application No. 19/251,066, published Oct. 16, 2025.

AWARDS & HONORS

  • 2025 Ph.D. Student Research Encouragement Fellowship Sep.2025-Aug.2026
    (2025년도 이공분야 학술연구지원사업(박사과정생연구장려금지원사업))

    • Funder: National Research Foundation of Korea (NRF)

    • Grant: 25,000,000 KRW

  • Dean's list, Outstanding grade (Kwangwoon university) [picture] Apr.2020

  • 4th place, Korea Intelligent Robot Contest (KIRC) [picture1][picture2] Oct.2019

  • Dean’s list , Outstanding grade (Kwangwoon university) [picture] Apr.2019

SKILLS

  • Programming Languages C/C++, Python, Java, LaTex

  • Machine Learning Library Pytorch

  • Research Skills Computer vision, Machine Learning, Multi-view geometry, Human pose and shape estimation, Representation Learning, 3D face reconstruction, 6DoF head pose estimation

Google Sites
Report abuse
Page details
Page updated
Google Sites
Report abuse