Younjoon Chung      

Email | CV | Google Scholar | Github | LinkedIn

I am a Research Associate at the Human Sensing Lab at Carnegie Mellon University. My research interests lie in the intersection of machine learning, computer vision and healthcare. Specifically, I focus on developing responsible AI systems towards equitable healthcare.

Previously, I was an M.S. in Computer Vision (MSCV) student in the Robotics Institute at Carnegie Mellon University, advised by Prof. Fernando De la Torre. Before CMU, I was a research engineer at VUNO Inc., where I developed algorithms for mobile digital radiography devices and computer-aided detection (CADe) systems. I also worked on various medical imaging projects with Prof. Yoon-Chul Kim at Samsung Medical Center.

In my free time, I enjoy making contributions to open-source projects, playing tennis, and watching baseball. I led the official Korean translation of the Keras documentation project, and have over 25+ commits merged in the Keras ecosystem.

Recent Highlights
My paper "Domain Gap Embeddings for Generative Dataset Augmentation" has been accepted to CVPR 2024.
Joined CMU School of Computer Science, Robotics Institute as a Research Associate, Spring 2024.
Graduated from CMU M.S. in Computer Vision in Dec 2023, advised by Prof. Fernando De la Torre.
Submitted a paper in CVPR 2024, Nov 2024. Fingers crossed!
Published a paper in Physica Medica, with Prof. Yoon-Chul-Kim, Nov 2023. [Link]
Joined CMU as an M.S. in Computer Vision student, Aug 2022.
Publications
Domain Gap Embeddings for Generative Dataset Augmentation
Younjoon Chung*, Oliver Wang*, Chen Wu, Fernando De la Torre
CVPR 2024  
Few-shot, fine-tuning free pipeline for targeted dataset generation.
Deep learning for classification of late gadolinium enhancement lesions based on the 16-segment left ventricular model
Yoon-Chul Kim, Younjoon Chung, Yeon Hyeon Choe
Physica Medica (2023)  
Paper
Segmental analysis of myocardial late gadolinium enhancement (LGE) lesions.
Automatic localization of anatomical landmarks in cardiac MR Perfusion using random forests
Yoon-Chul Kim, Younjoon Chung, Yeon Hyeon Choe
Biomedical Signal Processing and Control (2017)  
Paper
Learning-based approach to distinguish landmarks and non-landmarks for LV center and RV insertion point.
EVCMR: A tool for the quantitative evaluation and visualization of cardiac MRI data
Yoon-Chul Kim, Khu Rai Kim, Kwanghee Choi, Minwoo Kim, Younjoon Chung, Yeon Hyeon Choe
Computers in Biology and Medicine (2019)  
Paper
Open-source CMR post-processing tool for evaluating cardiac MRI data.
Patents
Method to read chest image
Beomhee Park, Minki Chung, Seo Taek Kong, Younjoon Chung
U.S. Patent Application 17/466,697, filed March 10, 2022.
The method includes: determining whether or not to identify presence of cardiomegaly for a chest image; detecting a lung region and a heart region respectively, and calculating a cardiothoracic ratio of the chest image using the detected lung region and the detected heart region.
Method for detecting abnormal findings and generating interpretation text of medical image
Minki Chung, Beomhee Park, Seo Taek Kong, Younjoon Chung
U.S. Patent Application 17/471,001, filed March 17, 2022.
The method includes: generating anatomical location information for the one or more lesions using an anatomical analysis model, and generating a diagnosis result for the received body medical image.
Miscellaneous
  • Mentor, CMU Undergraduate AI Mentoring Program (Fall 2023, Spring 2024)
  • Committee Member, Keras Korea (Mar 2018 - Present)
  • Technology Evangelist, Microsoft Student Partners (Mar 2015 - Mar 2016)
Template adopted from: 1 and 2