Hakmin Lee

Research Interests

• Machine Learning

Deep Learning

eXplainable AI

• Computer Vision

Image and Video Processing

Generative Adversarial Networks

Adversarial Attack&Defense

Out-of-distribution Detection

Education

• KAIST, Daejeon, Korea (2016 - Present)

Ph.D. in Electrical Engineering

· Advisor: Prof. Yong Man Ro

• KAIST, Daejeon, Korea (2016)

B.S. in Electrical Engineering

Publications

1. Feature2mass: Visual feature processing in latent space for realistic labeled mass generation

  Jae-Hyeok Lee, Seong Tae Kim, Hakmin Lee, Yong Man Ro

     Proceedings of the European Conference on Computer Vision (ECCV) Workshops, 2018

2. ICADx: interpretable computer aided diagnosis of breast masses

     {Hakmin Lee*, Seong Tae Kim*}, Hak Gu Kim, Yong Man Ro  (* Co-First Authors)

     SPIE Medical Imaging, 2018

3. Visually interpretable deep network for diagnosis of breast masses on mammograms

Seong Tae Kim, Jae-Hyeok Lee, Hakmin Lee, Yong Man Ro

Physics in Medicine & Biology (PMB), 2018

4. Lightweight and effective facial landmark detection using adversarial learning with face geometric map generative network

Hong Joo Lee, Seong Tae Kim, Hakmin Lee, Yong Man Ro

IEEE Transactions on Circuits and Systems for Video Technology (TCSVT), 2019

5. Realistic breast mass generation through BIRADS category

Hakmin Lee, Seong Tae Kim, Jae-Hyeok Lee, Yong Man Ro

International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2019

6. Robust ensemble model training via random layer sampling against adversarial attack

{Hakmin Lee*, Hong Joo Lee*}, Seong Tae Kim, Yong Man Ro (* Co-First Authors)

British Machine Vision Conference (BMVC), 2020

7. Revisiting role of autoencoders in adversarial settings

Byeong Cheon Kim, Jung Uk Kim, Hakmin Lee, Yong Man Ro

IEEE International Conference on Image Processing (ICIP), 2020

8. Defending Person Detection Against Adversarial Patch Attack by using Universal Defensive Frame

Youngjoon Yu*, Hong Joo Lee*, Hakmin Lee and Yong Man Ro

IEEE Transactions on Image Processing (TIP), 2022 (Accepted, IF:11.041 )

9. Adversarial anchor-guided feature refinement for adversarial defense

Hakmin Lee, Yong Man Ro

Image and Vision Computing (IMAVIS), 2023

Patents

1. Method for generating breast masses according to the lesion characteristics and the system thereof

Korea Patent Reg. 10-2067340

2. Interpreting method for diagnostic decision of deep network using breast imaging-reporting and data system and the system thereof

Korea Patent Reg. 10-2223255

3. Interactive computer-aided diagnosis method for lesion diagnosis and the system thereof

Korea Patent Reg. 10-2281988

4. White pattern defense apparatus and method against adversarial attack

Korea Patent Reg. 10-2491609

Awards & Honors

• Robert F. Wagner All Conference Best Student Paper Award, SPIE Medical Imaging (2018)

Best student paper award in computer-aided diagnosis conference (where there were 140 presentations)

Selected as final 9 papers over all conferences (9 papers are selected among 900 presentations)

• National Government Fellowship (2017 - Present)

Government of South Korea

Teaching

• EE667 Multiple View Geometry, KAIST (2020)

Teaching Assistant

• EE636 Digital Video Processing, KAIST (2018, 2019)

Teaching Assistant