Tae Soo Kim (김태수)

I am a research scientist and a team lead at Lunit. I work on research and development of deep learning models for medical image analysis to improve cancer patient care. Currently, I am building CV/ML systems for applications in cancer screening.

In summer of 2021, I defended my PhD in Computer Science at Johns Hopkins University where I developed deep learning models for recognizing fine-grained interactions from videos with my primary advisor Prof. Gregory Hager . I also worked closely with Prof. Alan Yuille and Austin Reiter.

I received both B.S and M.S.E in computer science from JHU. I am from Seoul, Korea and I enjoy playing golf. I also served as the 9-th president of the Korean Graduate Student Association at JHU.

CV  /  Google Scholar  /  Github

profile photo

  • 2023-06: One paper "Enhancing Breast Cancer Risk Prediction by Incorporating Prior Images" early accepted to MICCAI 2023.
  • 2022-07: One paper "OOOE: Only-One-Object-Exists Assumption to Find Very Small Objects in Chest Radiographs" accepted as an oral to MICCAI AMAI Workshop 2022.
  • 2022-06: One paper "Did You Get What You Paid For? Rethinking Annotation Cost of Deep Learning Based Computer Aided Detection in Chest Radiographs" accepted to MICCAI 2022.
  • 2022-05: One paper "Video-based assessment of intraoperative surgical skill" accepted to IJCARS 2022.
  • 2022-02: Our MICCAI tutorial on AI for medical image analysis in practice is accepted. See you at MICCAI!
  • 2021-11: One workshop paper "Learning from synthetic vehicles" accepted to WACV-2022 RWS workshop.
  • 2021-08: One paper "Motion Guided Attention Fusion to recognize interactions from videos" accepted to ICCV-2021.
  • 2021-06: I joined Lunit Inc. as a full-time research scientist!
  • 2021-05: I successfully defended my PhD thesis "Model-driven and Data-driven Methods for Recognizing Compositional Interactions from Videos".


I'm interested in computer vision and deep learning. Much of my research is about fine-grained recognition and learning with little to no labeled data. Currently, I am most excited about data-centric AI model development, especially the role of data quality in building CV applications in practice.

Did You Get What You Paid For? Rethinking Annotation Cost of Deep Learning Based Computer Aided Detection in Chest Radiographs
Tae Soo Kim*, Geonwoon Jang*, Sanghyup Lee, Thijs Kooi
MICCAI, 2022

Do we really know how much annotated data we need to reach a certain desired computer aided diagnosis (CAD) system performance? We define the cost of building a deep-learning based CAD system with respect to the following three dimensions: Granularity, Quantity and Quality of annotations. We investigate how each dimension ultimately impacts the resulting CAD performance and provide guidance to practitioners on how to optimize for data cost when building CAD systems for chest radiographs.

Video-based assessment of intraoperative surgical skill
Sanchit Hira, Digvijay Singh, Tae Soo Kim, Shobhit Gupta, Gregory D. Hager, Shameema Sikder, Swaroop Vedula, Gregory D. Hager
IJCARS, 2022

Can deep learning models assess the quality of a surgery directly from a video? We show that our video analysis model can accurately assess surgical skill from real world cataract surgeries.

Learning from synthetic vehicles
Tae Soo Kim, Bohoon Shim, Michael Peven, Weichao Qiu, Alan Yuille, Gregory D. Hager
WACV-RWS, 2022 / Dataset publicly available here

Simulated Articulated VEhicles Dataset (SAVED) is the first dataset of synthetic vehicles with moveable parts. Using SAVED, we show that we can train a model with synthetic images to recognize fine-grained vehicle parts and orientation directly from real images.

Motion Guided Attention Fusion to recognize interactions from videos
Tae Soo Kim, Jonathan Jones, Gregory D. Hager
ICCV, 2021

Do current video models have the ability to recognize an unseen instantiation of an interaction defined using a combination of seen components? We show that it ispossible by specifying the dynamic structure of an action using a sequence of object detections in a top-down fashion. When the top-down structure is combined with a dual-pathway bottom-up approach, we show that the model can then generalize even to unseen interactions.

DASZL: Dynamic Action Signatures for Zero-shot Learning
Tae Soo Kim*, Jonathan Jones*, Michael Peven*, Zihao Xiao, Jin Bai, Yi Zhang, Weichao Qiu, Alan Yuille, Gregory D. Hager
AAAI, 2021

This compositional approach allows us to reframe fine-grained recognition as zero-shot activity recognition, where a detector is composed “on the fly” from simple first-principles state machines supported by deep-learned components. Listen to Dr. Alan Yuille talk about this work here (from 15:00 and on)!

SAFER: Fine-grained activity detection by compositional hypothesis testing
Tae Soo Kim, Yi Zhang, Zihao Xiao, Michael Peven, Weichao Qiu, Jin Bai, Alan Yuille, Gregory D. Hager
arxiv, 2019

SAFER models a large space of fine-grained activities using a small set of detectable entities and their interactions. Such a design scales effectively with concurrent developments of object detectors, parsers and more. Our model effectively detects fine-grained human activities without any activity level supervision in video surveillance applications.

Synthesizing attributes with unreal engine for fine-grained activity analysis
Tae Soo Kim, Michael Peven, Weichao Qiu, Alan Yuille, Gregory D. Hager
WACV-W, 2019

Recent deep neural network based computer vision models can be trained to recognize pretty much anything given enough data. We show we can synthesize visual attributes using the UnrealEngine4 to train activity classification models.

Assessment of automated identification of phases in videos of cataract surgery using machine learning and deep learning techniques
Tae Soo Kim*, Felix Yu*, Gianluca Silva Croso*, Ziang Song, Felix Parker, Gregory D. Hager, Austin Reiter, Swaroop Vedula, Haider Ali, Shameema Sikder
JAMA Network Open, 2019

Competence in cataract surgery is a public health necessity, and videos of cataract surgery are routinely available to educators and trainees but currently are of limited use in training. We develop tools that efficiently segment videos of cataract surgery into constituent phases for subsequent automated skill assessment and feedback.

Objective assessment of intraoperative technical skill in capsulorhexis using videos of cataract surgery
Tae Soo Kim, Molly O'Brien, Sidra Zafar, Anand Malpani, Gregory D. Hager, Shameema Sikder, Swaroop Vedula
IJCARS, 2019

We introduce a model for objective assessment of surgical skill from videos of microscopic cataract surgery. Our model can accurately predict surgeon's skill level from tool tip movements captured in the surgical view.

Crowdsourcing annotation of surgical instruments in videos of cataract surgery
Tae Soo Kim, Anand Malpani, Austin Reiter, Gregory D. Hager, Shameema Sikder, Swaroop Vedula
MICCAI-W, 2018

We evaluate reliability and validity of crowdsourced annotations for information on surgical instruments (name of instruments and pixel location of key points on instruments) in cataract surgery.

UnrealCV: Virtual Worlds for Computer Vision
Weichao Qiu, Fangwei Zhong, Yi Zhang, Siyuan Qiao, Zihao Xiao, Tae Soo Kim, Yizhou Wang, Alan Yuille
ACM-MM, 2017

UnrealCV is an open source project to help computer vision researchers build virtual worlds using Unreal Engine 4 (UE4). Check it out here.

Interpretable 3D Human Action Analysis with Temporal Convolutional Networks
Tae Soo Kim, Austin Reiter
CVPR-W, 2017

We re-design the TCN with interpretability in mind and take a step towards a spatio-temporal model that is easier to understand, explain and interpret.

Reviewer, CVPR, ICCV, ECCV

Reviewer, AAAI

jhu Head Teaching Assistant for EN.600.661, Computer Vision. Fall 2015, Fall 2016

Head Teaching Assistant for EN.600.684, Augmented Reality. Spring 2016

Head Teaching Assistant for EN.600.107, Introductory Programming in Java. Summer 2015

Head Teaching Assistant for EN.600.226, Data Structures. Spring 2015

Website template credits
Also, consider using Leonid Keselman's Jekyll fork of this page.