Biography
Hyeonwoo Cho is an AI researcher at OSSTEM IMPLANT. He received his bachelor’s and master’s degrees from Kyushu University. Additionally, I have worked as an AI researcher at VUNO. My research interests include Domain Adaptation and 3D Vision.
Interests
- Domain Adaptation
- 3D vision
- Noisy Label Learning
- Medical Imaging
- Self-Supervised Learning
- Semi-Supervised Learning
- MultiModal
Education
MS in Information Science and Electrical Engineering, 2022
Kyushu University
BS in Aeronautics and Astronautics, 2020
Kyushu University
Skills
Technical
Python
SQL
Pytorch
Pytorch Lightning
Docker
Git
Hobbies
Soccer
Travel
Running
Experience
AI Research Engineer (Alternative Military Service)
OSSTEM IMPLANT
AI Developement Team:
- Developed A RAG System for Answering about Dental-Realated Questions. (AI chatbot)
- Developing A Crown Geneeration Model from 3D Mesh and Point-Cloud. (Graphics & Generation)
- Developing An Automated Orthodontic Treatment Planning System. (Graphics & Generation)
AI Research Scientist (Alternative Military Service)
VUNO
BLU3 Team (3D Domain):
- Developed A Early Dignosis AI system for Lung Cancer. (3D Lung CT)
- Ansysis of Frontotemporal dementia (FTD) based Brain Volumetric Information. (3D Brain MRI)
MS in Computer Vision
Kyushu Univ.
Human Interface Lab:
- Advisor: Ryoma Bise
- Topic: Object Detection, Tracking and Domain Adaptation
Recent News
- My one paper has been accepted in ACCV 2024
Publications
(2024).
Automated Brain Volumetry Analysis for Differential Diagnosis of Frontotemporal Dementia Subtypes.
Alzheimer’s Association International Conference.
(2024).
CNG-SFDA: Clean-and-Noisy Region Guided Online-Offline Source-Free Domain Adaptation.
Proceedings of the Asian Conference on Computer Vision.
(2024).
Comparison of intracranial volume adjustment methods to evaluate brain atrophy severity in AD continuum.
Alzheimer’s Association International Conference.
(2024).
Joint-Embedding Predictive Architecture for Self-Supervised Learning of Mask Classification Architecture.
arXiv preprint arXiv:2407.10733.
(2024).
Quantitative Analysis of Choroid Plexus Enlargement in Alzheimer's Dementia: A Study of Automated Volumetric Technique.
Alzheimer’s Association International Conference.
Projects
*
In this project, we build AI that generates a crown mesh from neighborhood tooth mesh. - Deep Learning
A RAG system for answering questions about dental-related question.
The Isolated Sign Language Recognition competition’s goal is to classify isolated American Sign Language (ASL) signs. This is my Silver Medal solution.
A system for detecting and analyzing lung cancer risk factors from lung CTs of patients with lung diseases.
