Welcome

I am a Ph.D. candidate in Robotics and Computer Vision Lab. at KAIST under the supervision of Prof. In So Kweon Ph.D. since 2016. I received the B.Sc. degree in electrical and computer engineering from UNIST, South Korea and the M.Sc. degree in the robotics program from KAIST under the supervision of Prof. David Hyunchul Shim Ph.D. in 2015.

I was a research intern in Visual Computing Group at Microsoft Research Asia (MSRA) during 2018, working with Dr. Stephen Lin.

Research Interests

"Learning Semantic and Geometric Perception for Robust Autonomous Driving"

  • Computer Vision
    • Object detection/recognition & segmentation
    • Depth & pose & optical flow estimation
    • Geometric view synthesis
    • Sensor fusion
  • Deep Learning
    • Representation learning
    • Domain adaptation
    • Few-shot learning
    • Multi-modal attention

Preprints

Instance-wise Depth and Motion Learning from Monocular Videos

Seokju Lee, Sunghoon Im, Stephen Lin, In So Kweon

ArXiv preprint, Dec 2019

[Paper] [Project]

International Publications

2020

Unsupervised Intra-domain Adaptation for Semantic Segmentation through Self-Supervision

Fei Pan, Inkyu Shin, Francois Rameau, Seokju Lee, In So Kweon

IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), Jun 2020

[Paper] [Project] [Code] [Video]

Oral Presentation

2019

Learning Residual Flow as Dynamic Motion from Stereo Videos

Seokju Lee, Sunghoon Im, Stephen Lin, In So Kweon

IEEE International Conference on Intelligent Robots and Systems (IROS), Nov 2019

[Paper] [Project] [Video]

Oral Presentation

Visuomotor Understanding for Representation Learning of Driving Scenes

Seokju Lee, Junsik Kim, Tae-Hyun Oh, Yongseop Jeong, Donggeun Yoo, Stephen Lin, In So Kweon

The British Machine Vision Conference (BMVC), Sep 2019

[Paper] [Supp] [Dataset] [Poster] [Project] [Video]

Revisiting Residual Networks with Nonlinear Shortcuts

Chaoning Zhang, Francois Rameau, Seokju Lee, Junsik Kim, Philipp Benz, Dawit M. Argaw, Jean-Charles Bazin, In So Kweon

The British Machine Vision Conference (BMVC), Sep 2019

[Paper]

Spotlight Presentation

Variational Prototyping-Encoder: One-Shot Learning with Prototypical Images

Junsik Kim, Tae-Hyun Oh, Seokju Lee, Fei Pan, In So Kweon

IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), Jun 2019

[Paper] [Project]

2018

Co-domain Embedding using Deep Quadruplet Networks for Unseen Traffic Sign Recognition

Junsik Kim, Seokju Lee, Tae-Hyun Oh, In So Kweon

AAAI Conference on Artificial Intelligence (AAAI), Feb 2018

[Paper] [Project]

Received Best Poster Presentation Award, 30th IPIU 2018

2017

VPGNet: Vanishing Point Guided Network for Lane and Road Marking Detection and Recognition

Seokju Lee, Junsik Kim, Jae Shin Yoon, Seunghak Shin, Oleksandr Bailo, Namil Kim, Tae-Hee Lee, Hyun Seok Hong, Seung-Hoon Han, In So Kweon

IEEE International Conference on Computer Vision (ICCV), Oct 2017

[Paper] [Project] [Code] [Poster] [Patent]

Patent in Korea (10-2018-0048407)

Pixel-Level Matching for Video Object Segmentation Using Convolutional Neural Networks

Jae Shin Yoon, Francois Rameau, Junsik Kim, Seokju Lee, Seunghak Shin, In So Kweon

IEEE International Conference on Computer Vision (ICCV), Oct 2017

[Paper] [Project]

Robust Road Marking Detection and Recognition Using Density-Based Grouping and Machine Learning Techniques

Oleksandr Bailo, Seokju Lee, Francois Rameau, Jae Shin Yoon, In So Kweon

IEEE Winter Conference on Applications of Computer Vision (WACV), Mar 2017

[Paper] [Video]

2016

Fast multiple objects detection and tracking fusing color camera and 3D LIDAR for intelligent vehicles

Soonmin Hwang, Namil Kim, Yukyung Choi, Seokju Lee, In So Kweon

IEEE International Conference on Ubiquitous Robots and Ambient Intelligence (URAI), Aug 2016

[Paper] [Project]

2015

Rich feature hierarchies from omni-directional RGB-DI information for pedestrian detection

Seokju Lee, Sungsik Huh, Donggeun Yoo, In So Kweon, David Hyunchul Shim

IEEE International Conference on Ubiquitous Robots and Ambient Intelligence (URAI), Oct 2015

[Paper]

2014

Development of an exploration rover platform for sample return mission

Seokju Lee, Sungsik Huh, Sungyeon Park, David Hyunchul Shim

IEEE International Conference on Ubiquitous Robots and Ambient Intelligence (URAI), Nov 2014

[Paper] [Video]

Other Publications


  • A Flight Test for Neutralization of Enemy Position Using Intelligent Cooperative System of Heterogeneous Unmanned Vehicle System

Hanjun Song, Seokju Lee, Sungwook Cho, and David Hyunchul Shim

The Korean Society for Aeronautical and Space Sciences Spring Conference (KSAS), Apr 2015

[Paper] [IEEE Spectrum]


  • Development of Robotic Hand System Using Vision-Based Motion Sensor

Seokju Lee, Hanjun Song, and David Hyunchul Shim

The 29th Institute of Control, Robotics and Systems Annual Conference (ICROS), May 2014

[Paper] [Project] [Video]


  • Development of a Rover Platform for Exploration and Sample Return Mission

Seokju Lee, Sungsik Huh, Dasol Lee, Sungwook Cho, and David Hyunchul Shim

The Korean Society for Aeronautical and Space Sciences Spring Conference (KSAS), Apr 2014

[Paper] [Video]

Thesis


  • Hierarchical Pedestrian Detection Using Omnidirectional Camera and Laser Sensor Fusion

MS Thesis, 2015.

[Paper]

Patents


  • APPARATUS AND METHOD FOR DETECTING A LANE

KR 10-2018-0048407


  • Vanishing-Guided Lane and Road Marking Detection via Multi-Task Network

Application date: Oct. 31, 2016 (US Provisional Application 62/414,951)

Academic Activities


  • Program Committee

- 3DV 2020


  • Journal Reviewer

- Autonomous Robots (AURO)


  • Conference Reviewer

- CVPR 2018, 2019, 2020

- NeurIPS 2020

- WACV 2021

- ACCV 2020

Last updated: May 29, 2020