Jung Uk Kim


Assistant Professor at Kyung Hee University (Computer Science and Engineering)


Visual Artificial Intelligence Lab (https://visualai.khu.ac.kr)

E-mail: ju.kim [at] khu.ac.kr

[CV] [Google Scholar]

Hello! I am currently an Assistant Professor of Computer Science and Engineering at Kyung Hee University.

I graduated Ph. D. in Electrical Engineering at Korea Advanced Institute of Science and Technology (KAIST) under the supervision of Prof. Yong Man Ro. My research interests include machine learning and computer vision. In particular, my research topics include object detection in challenging environments (e.g., small-scale, occlusion, weather, etc.), pedestrian detection, and multimodal (multispectral) object detection. Also, I am interested in semantic segmentation and object tracking.

Education

  • Ph.D. KAIST (Mar. 2018 - Aug. 2022)

School of Electrical Engineering, KAIST, Daejeon, KoreaAdvisor: Prof. Yong Man Ro
  • M.S. KAIST (Mar. 2016 - Feb. 2018)

School of Electrical Engineering, KAIST, Daejeon, KoreaAdvisor: Prof. Yong Man Ro
  • B.S. Ajou University (Mar. 2012 - Feb. 2016) (Summa Cum Laude)

Electrical Engineering, Ajou, Suwon, Korea

Research Interests

  • Computer Vision

  • Machine Learning (Unsupervised Learning, Self-supervised Learning)

  • Visual Recognition - 2D/3D/Video Object Detection, Segmentation

  • Multi-modal AI - Audio-Visual Model, Multispectral (RGB/IR) Model

  • Probabilistic AI - Bayesian Modeling, Uncertainty

  • Domain Adaptation

  • Generative Adversarial Network (GAN)

But not limited to

News

[Jan. 2022] One paper is accepted in ICASSP 2022

[Dec. 2021] One paper is accepted in AAAI 2022 Oral Presentation

2021: ICCV (1 paper), IEEE TCSVT (1 paper), ICASSP (1 paper) are accepted

2020: CVPR (1 paper), ECCV (1 paper), IEEE TCSVT (1 paper), ICASSP (1 paper), ICIP (6 papers) are accepted

2019: IEEE TCSVT (1 paper), ICIP (1 paper) are accepted

Publications (Selected)

2022

  • Uncertainty-Guided Cross-Modal Learning for Robust Multispectral Pedestrian Detection

Jung Uk Kim, Sungjune Park, and Yong Man RoIEEE Transactions on Circuits and Systems for Video Technology (TCSVT), 2022 [pdf]
  • Robust Thermal Infrared Pedestrian Detection by Associating Visible Pedestrian Knowledge

Sungjune Park*, Dae Hwi Choi*, Jung Uk Kim, and Yong Man Ro (*Both authors equally contributed)IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2022 [pdf]
  • Towards Versatile Pedestrian Detector with Multisensory-Matching and Multispectral Recalling Memory

Jung Uk Kim, Sungjune Park, and Yong Man RoAssociation for the Advancement of Artificial Intelligence (AAAI), 2022 (Oral Presentation) (Acceptance rate : 4.9%) [pdf]

2021

  • Robust Small-scale Pedestrian Detection with Cued Recall via Memory Learning

Jung Uk Kim*, Sungjune Park*, and Yong Man Ro (*Both authors equally contributed)IEEE International Conference on Computer Vision (ICCV), 2021 [pdf] (Acceptance rate : 25.9%)
  • CUA Loss: Class Uncertainty-Aware Gradient Modulation for Robust Object Detection

Jung Uk Kim, Seong Tae Kim, Hong Joo Lee, Sangmin Lee, and Yong Man RoIEEE Transactions on Circuits and Systems for Video Technology (TCSVT), 2021 [pdf]
  • Towards Robust Training of Multi-Sensor Data Fusion Network Against Adversarial Examples in Semantic Segmentation

Youngjoon Yu, Hong Joo Lee, Byeong Cheon Kim, Jung Uk Kim, and Yong Man RoIEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2021 [pdf]

2020

  • SACA Net: Cybersickness Assessment of Individual Viewers for VR Content via Graph-based Symptom Relation Embedding

Sangmin Lee, Jung Uk Kim, Hak Gu Kim, Seongyeop Kim, and Yong Man RoEuropean Conference on Computer Vision (ECCV), 2020 [pdf] (Acceptance rate : 27.1%)
  • Structure Boundary Preserving Segmentation for Medical Image with Ambiguous Boundary

Hong Joo Lee, Jung Uk Kim, Sangmin Lee, Hak Gu Kim, and Yong Man RoIEEE Computer Vision and Pattern Recognition (CVPR), 2020 [pdf] (Acceptance rate : 22.1%)
  • BBC Net: Bounding-Box Critic Network for Occlusion-Robust Object Detection

Jung Uk Kim, Jungsu Kwon, Hak Gu Kim, and Yong Man RoIEEE Transactions on Circuits and Systems for Video Technology (TCSVT), 2020 [pdf]
  • Towards High-Performance Object Detection: Task-Specific Design Considering Classification and Localization Separation

Jung Uk Kim, Seong Tae Kim, Eun Sung Kim, Sang-Keun Moon, and Yong Man RoIEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2020 [pdf]
  • Towards Human-Like Interpretable Object Detection via Spatial Relation Encoding

Jung Uk Kim*, Sungjune Park*, and Yong Man Ro (*Both authors equally contributed)IEEE International Conference on Image Processing (ICIP), 2020 [pdf]
  • Class Incremental Learning with Task-Selection

Eun Sung Kim*, Jung Uk Kim*, Sangmin Lee, Sang-Keun Moon, and Yong Man Ro (*Both authors equally contributed)IEEE International Conference on Image Processing (ICIP), 2020 [pdf]
  • Comprehensive Facial Expression Synthesis using Human-Interpretable Language

Joanna Hong, Jung Uk Kim, Sangmin Lee, and Yong Man RoIEEE International Conference on Image Processing (ICIP), 2020 [pdf]
  • Fake Video Detection With Certainty-Based Attention Network

Dae Hwi Choi, Hong Joo Lee, Sangmin Lee, Jung Uk Kim, and Yong Man RoIEEE International Conference on Image Processing (ICIP), 2020 [pdf]

2019

  • Attentive Layer Separation for Object Classification and Object Localization in Object Detection

Jung Uk Kim and Yong Man RoIEEE International Conference on Image Processing (ICIP), 2019 [pdf]

2018

  • Object Bounding-Box-Critic Networks for Occlusion Robust Object Detection in Road Scene

Jung Uk Kim*, Jungsu Kwon*, Hak Gu Kim, Haesung Lee, and Yong Man Ro (*Both authors equally contributed)IEEE International Conference on Image Processing (ICIP), 2018 [pdf]

2017

  • Robust and Real-Time Visual Tracking with Triplet Convolutional Neural Network

Jung Uk Kim, Hak Gu Kim, and Yong Man RoACM Multimedia (ACM MM) Workshop, 2017 [pdf]
  • Iterative Deep Convolutional Encoder-Decoder Network for Medical Image Segmentation

Jung Uk Kim, Hak Gu Kim, and Yong Man RoInternational Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2017 [pdf]
  • Object Tracking based on Weight Sharing CNN Structure According to Search Area Setting Method Considering Object Movement

Jung Uk Kim, Hak Gu Kim, and Yong Man RoJournal of Korea Multimedia Society, 2017

Awards & Honors

  • Certificate of Appointment of Korean Intellectual Property Office Advisor

  • Best Paper Award

    • Korea Signal Processing Conference, 2019
    • Korea Multimedia Society, 2017
  • Scholarships

    • AAAI-22 Student Scholar Program (from IBM)
    • Korean Government Scholarships, KAIST (2016-2018)

Experiences

  • Korean Intellectual Property Office Advisor (Apr. 2019 - Present)

[Teaching Assistant]

    • Digital Video Processing (2017, Fall): KAIST

    • Introduction to Multimedia (2017, Spring): KAIST

Academic Activities

[Reviews]

  • International Conference on Computer Vision (ICCV) 2021

  • Conference on Computer Vision and Pattern Recognition (CVPR) 2021, 2022

  • European Conference on Computer Vision (ECCV) 2022

  • ACM Multimedia (ACM MM) 2018 Workshop

  • Reviewer for Peer-Review Journals

    • IEEE Transactions on Medical Imaging (TMI)

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

    • IEEE Signal Processing Letters (SPL)

    • IEEE Robotics and Automation Letters (RA-L)

    • International Journal of Remote Sensing (IJRS)

[Talks]

  • Deep Learning Technology in Audio and Image

Korean Intellectual Property Office, 2020
  • The Present and Future of Deep Learning

Korean Intellectual Property Office, 2019

[Organization]

  • Organizing Chair at Workshop for Center for Applied Research in Artificial Intelligence (CARAI) 2021