Donghyun Kim
Scholar

Donghyun Kim

Google Scholar ID: u9bQ6acAAAAJ
Yonsei University
3D Vision3D Point CloudImage Restoration
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
293
 
H-index
3
 
i10-index
1
 
Publications
4
 
Co-authors
0
 
Contact
CVOpen ↗GitHubOpen ↗LinkedInOpen ↗
Publications
2 items
Rethinking Graph Convolution for 2D-to-3D Hand Pose Lifting
2026
Cited
0
PAC-BENCH: Evaluating Multi-Agent Collaboration under Privacy Constraints
2026
Cited
0
Resume
Academic Achievements
  • Publications:
  • - PLATYPUS: Progressive local surface estimator for arbitrary-scale point cloud upsampling, AAAI 2025 (Oral Presentation, top 4.6%)
  • - Fourier Decomposition for Explicit Representation of 3D Point Cloud Attributes, Preprint
  • - FALCON: Fast Image Haze Removal Leveraging Continuous Density Mask, CVPR 2025 Workshop (NTIRE)
  • - WoLF: Large language model framework for CXR understanding, Preprint
  • Awards & Honors:
  • - 2nd Prize, Coding Challenge, Korea Institute of Startup & Entrepreneurship Development (KISED), 2021
  • - Grand Prize, Software Capstone Project, Yonsei University, 2023
  • - Exhibition Popularity Award, Software Capstone Project, Yonsei University, 2022
Research Experience
  • MICV Lab, Yonsei University, Seoul, Jul 2022 - Present, Graduate Student Researcher, focusing on 3D Point Cloud, Image Restoration, and Medical Imaging
  • DC Lab, Yonsei University, Seoul, Jan 2021 - Jun 2021, Undergraduate Research Intern, Project with Severance Children's Hospital
  • Arizona State University, Tempe, AZ, Feb 2024, Exchange Program, research areas include Image Restoration, Domain Generalization, and Medical Imaging
  • Innoyard, Seoul, Jun 2021 - Aug 2022, Researcher and Software Engineer, working on 3D Data Processing (Point Cloud Registration & Surface Reconstruction) and Development of Mobile and Tablet PC Application (Flutter & Swift)
  • Linz Private Equity, Seoul, Jan 2020 - Feb 2020, Intern, Stock Data Analysis
Education
  • M.S./Ph.D. in Computer Science, Yonsei University, Sep 2023 - Present, Advisor: Professor Seong Jae Hwang
  • B.S. in Computer Science, Yonsei University, Mar 2019 - Aug 2023
Background
  • Research interests: 3D vision, with a particular focus on 3D point clouds; enhancing the quality of 3D data from a low-level vision perspective and exploring encoding approaches that are well-suited to the unique characteristics of 3D data.
Miscellany
  • Looking for internship opportunities and open to research collaboration, especially in 3D vision.