Donggeun Yoo

Co-founder & Chief of Research at Lunit
dgyoo@lunit.io
Google Scholar | CV | Linkedin



Bio

I received BS in 2011, MS in 2013 and PhD in 2019 at School of Electrical Engineering form KAIST, South Korea. The title of my dissertation (composed of four chapters, all published) was: Deep Learning Based Visual Recognition Robust Against Background Clusters, written under the supervision of Prof. In So Kweon. During the PhD course, I co-founded Lunit, a Seoul-based medical AI startup, with my lifelong friends in 2013. As the Chief of Research at Lunit, I am devoted to developing advanced medical AI for radiology and oncology. During my internship experience at Adobe Research in the US, I worked on large-scale video representation learning. My research interests include visual recognition problems with deep learning approaches.



Research Interests

Machine Learning 

Deep learning, un/semi-supervised learning, representation learning, active learning, domain generalization, large-scale learning method

Computer Vision 

Visual recognition, image classification, object detection, semantic segmentation, image retrieval, medical image analysis, data-driven imaging biomarker (DIB), computer vision for scientific discovery



Technical Achievements

Nov. 2019. 

Visual Domain Adaptation Challenge (VisDA) in ICCV 2019
Team Lunit won the 1st place in the semi-supervised domain adaptation task.
Method: Reducing Domain Gap via Style-Agnostic Networks.
Related paper | Workshop

Mar. 2017. 

My transfer learning method, Multi-Scale Pyramid Pooling (MPP), was employed to Samsung Galaxy S8 Bixby Vision for fine-grained object classification and product retrieval.
Related paper

Dec. 2015. 

ImageNet Large Scale Visual Recognition Challenge (ILSVRC) in ICCV 2015
Team Lunit-KAIST won the 5th place at the classification and localization task among 23 participants including Google, Microsoft, Samsung and Qualcomm.
Invited to ILSVRC Workshop to provide a talk about "Multi-Class AttentionNet", which was selected as one of top 3 novel localization approaches.
Result | Workshop | Slide | Related paper 1 | Related paper 2

Feb. 2009. 

Grand Prize in KAIST Undergraduate Research Program.
Topic: Portable Noncontact Heartbeat Sensor Using LC Oscillation.
Advisor: Prof. Songcheol Hong.



Academic Activities

2017 - Present 

Reviewer in CVPR, ICCV, ECCV, and other conferences.

Feb. 2021. 

Invited talk at Image Processing and Image Understanding (IPIU)
Topic: Conquer Cancer with AI: Challenges and Limitations
Link

Nov. 2019. 

Invited talk at Annual Symposium of the Korea Endocrine Society 2019 (medical conference)
Topic: The Potential of AI in Medicine: From Diagnostic AI to Predictive Biomarker

Oct. 2019. 

Organizing an ICCV 2019 Workshop: Visual Recognition for Medical Images (VRMI’19)
Co-organizers: Dr. Hoo-Chang Shin (NVIDIA) and Prof. Kyunghyun Cho (NYU and FAIR)
Link

Oct. 2019. 

Invited talk at MICCAI 2019 Workshop: Medical Informatics in Medical Image Analytics (MIMIA’19)
Topic: Reducing Annotation Cost in Medical Image Analysis
Link

Apr. 2019. 

Invited talk at Korea International Gastric Cancer Week 2019 (medical conference)
Topic: The Potential of AI in Medicine: From Diagnostic AI to Predictive Biomarker
Link

Dec. 2015. 

Invited talk at ICCV 2015 Workshop: ImageNet and MS COCO Visual Recognition Challenges Joint Workshop (ILSVRC)
Topic: Multi-class AttentionNet
Link



Selected Publications

Hyeonseob Nam*, HyunJae Lee*, Jongchan Park, Wonjun Yoon, Donggeon Yoo
Reducing Domain Gap by Reducing Style Bias,
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021.
Oral presentation
Preprint


Minchul Kim*, Jongchan Park*, Seil Na, Chang Min Park, Donggeun Yoo
Learning Visual Context by Comparison,
European Conference on Computer Vision (ECCV), 2020.
Spotlight presentation
Paper | Supp


Jaehwan Lee, Donggeon Yoo, Jung Yin Huh, Hyo-Eun Kim,
Photometric Transformer Networks and Label Adjustment for Breast Density Prediction,
IEEE International Conference on Computer Vision (ICCV) VRMI Workshop, 2019.
Paper


Inwan Yoo, Donggeun Yoo, Kyunghyun Paeng,
PseudoEdgeNet: Nuclei Segmentation only with Point Annotations,
International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2019.
Oral presentation
Paper


Seokju Lee, Junsik Kim, Tae-Hyun Oh, Yongseop Jeong, Donggeun Yoo, Stephen Lin, In So Kweon,
Visuomotor Understanding for Representation Learning of Driving Scenes,
The British Machine Vision Conference (BMVC), 2019.
Paper



Donggeun Yoo, In So Kweon,
Learning Loss for Active Learning,
IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2019.
Oral presentation
Paper | Supp



Jongchan Park, Joon-Young Lee, Donggeun Yoo, In So Kweon,
Distort-and-Recover: Color Enhancement using Deep Reinforcement Learning,
IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2018.
Paper | Supp | Code



Dahun Kim, Donghyeon Cho, Donggeun Yoo, In So Kweon,
Learning Image Representations by Completing Damaged Jigsaw Puzzles,
IEEE Winter Conference on Applications of Computer Vision (WACV), 2018.
Paper



Dahun Kim, Donghyeon Cho, Donggeun Yoo, In So Kweon,
Two-Phase Learning for Weakly Supervised Object Localization,
IEEE International Conference on Computer Vision (ICCV), 2017.
Paper



Donggeun Yoo, Sunggyun Park, Kyunghyun Paeng, Joon-Young Lee, In So Kweon,
Action-Driven Object Detection with Top-Down Visual Attentions,
arXiv preprint, 2016.
Paper | Video - person, multi



Donggeun Yoo, Namil Kim, Sunggyun Park, Anthony S Paek, In So Kweon,
Pixel-Level Domain Transfer,
European Conference on Computer Vision (ECCV), 2016.
Paper | Supp | LookBook dataset | Code written by Fei Xia
(This work was invited to TASK-CV)



Donggeun Yoo, Sunggyun Park, Joon-Young Lee, Anthony S Paek, In So Kweon,
AttentionNet: Aggregating Weak Directions for Accurate Object Detection,
IEEE International Conference on Computer Vision (ICCV), 2015.
Paper | Supp



Youngjin Yoon, Hae-Gon Jeon, Donggeun Yoo, Joon-Young Lee, In So Kweon,
Learning a Deep Convolutional Network for Light-Field Image Super-Resolution,
IEEE International Conference on Computer Vision (ICCV) - CPCV Workshop, 2015.
Paper



Donggeun Yoo, Sunggyun Park, Joon-Young Lee, In So Kweon,
Multi-scale Pyramid Pooling for Deep Convolutional Representation,
IEEE International Conference on Computer Vision and Pattern Recognition (CVPR) - DeepVision Workshop, 2014.
Paper | Supp
(Employed to Samsung Galaxy S8 Bixby Vision)



Donggeun Yoo, Kyunghyun Paeng, Sunggyun Park, Jungin Lee, Seungwook Paek, Sung-Eui Yoon, In So Kweon,
PRISM: A System for Weighted Multi-Color Browsing of Fashion Products,
International World Wide Web Conference Companion (WWW), 2014.
Paper