I am a research engineer. I received M.S. degrees from the Yonsei University, where I was advised by Ha Young Kim. I am broadly interested in machine learning, computer vision and multi modal models. Specifically, my research interests include

  • Object Recognition
  • Domain Generalization and Adaptation
  • Vision and Language(multi-modal) System

but not limited to.

🔥 News

  • 2023.02:  🎉🎉 One paper has been accepted at CVPR 2023.

📝 Publications

  • Decompose, Adjust, Compose: Effective Normalization by Playing with Frequency for Domain Generalization

    Computer Vision and Pattern Recognition (CVPR) 2023
    Sangrok Lee*, Jongseong Bae*, Ha Young Kim
    *:equal contribtion
    [paper][code]

  • Multi-attention-based soft partition network for vehicle re-identification

    Journal of Computational Design and Engineering (JCDE), 2023 SCIE(IF:6.16)
    Sangrok Lee, Taekang woo, Sang Hun Lee
    [paper][code]

  • ESGCN: Edge Squeeze Attention Graph Convolutional Network for Traffic Flow Forecasting

    Tech Report, 2022
    Sangrok Lee, Ha Young Kim
    [paper][code]

  • SBNet: Segmentation-based Network for Natural Language-based Vehicle Search

    Computer Vision and Pattern Recognition Workshop(CVPRW) 2021
    Sangrok Lee, Taekang woo, Sang Hun Lee
    [paper][code]

  • StRDAN: Synthetic-to-Real Domain Adaptation Network for Vehicle Re-Identification

    Computer Vision and Pattern Recognition Workshop(CVPRW) 2020
    Sangrok Lee, Eunsoo Park, Hongsuk Yi, Sang Hun Lee
    [paper] [code]

  • An Energy and GPU-Computation Efficient Backbone Network for Real-Time Object Detection

    Computer Vision and Pattern Recognition Workshop (CVPRW) 2019
    Youngwan Lee, Joongwon Hwang, Sangrok Lee, Yuseok Bae, Jongyoul Park
    [paper][timm]

📖 Educations

  • 2021.07 - 2023.08, Graduate School of Information, Yonsei University. Advised by Ha Young Kim.
  • 2012.03 - 2019.02, Hankuk University of Foreign Studies.

💬 Academic Services

Reviewer

  • IEEE / CVF Conference on Computer Vision and Pattern Recognition (CVPR): 2023
  • IEEE International Conference on Advanced Video and Signal-based Surveillance (AVSS): 2023

Pet Project

  • 2015 - Current, HUFSPOON (Full-Stack Software Engineer)

💻 Work Experiences