Home


Seokeon Choi (최석언)
Ph.D Candidate
Republic of Korea


Contact

E-mail : seokeon (at) kaist (dot) ac (dot) kr / bismex (at) gmail (dot) com
Tel : +82-42-350-7521
YouTube : https://www.youtube.com/channel/UC3s7aNMmUzYpfJNBZXolpGQ


Research interests
  • Object tracking (Long-term tracking)
  • Biometrics (Gait recognition, Person re-identification)
  • Image processing 
  • 3D vision 


Education

Project experience
  • 2018.02 ~ 2020.01
    • Hanwha Systems
    • Title : AI-based object tracking and recognition
    • Keyword : IR-based object tracking / Long-term tracking / Deep learning / Image recognition
  • 2017.09 ~ 2018.02
  • 2017.03 ~ 2017.07
    • Samsung Advanced Institute of Technology (SAIT)
    • Title : IR image quality assessment and score-level fusion for improving face recognition performance
    • Keyword : Image quality assessment (IQA) / Face recognition / Score-level fusion / Adaptive learning
  • 2016.03 ~ 2017.02
    • Samsung DMC
    • Title : Development of super-resolution technology in video data streaming environment
    • Keyword : Super-resolution with metadata / Texture transfer / Video streaming
  • 2015.03 ~ 2016.02
    • Samsung DMC
    • Title : Perception-based high-resolution creation
    • Keyword : Texture classification / Texture synthesizability / Super-resolution


Publication 
  • Seokeon Choi, Jonghee Kim, and Changick Kim, "Robust Model-based Gait Recognition via Frame-level Score Fusion," Submitted to IEEE Transactions on Information Forensics and Security, Jun. 2018.
    (Impact factor: 5.824)
  • Hankyeol Lee*, Seokeon Choi*, and Changick Kim, "A Memory Model based on the Siamese Network for Long-term Tracking", in Proc. European Conference on Computer Vision Workshop (ECCVW), Munich, Germany, Sep. 8-14, 2018. 
    (* These two authors contributed equally), [3rd prize winner at the VOT-LT2018 challenge]
  • Hankyeol Lee, Seokeon Choi, and Changick Kim, "A Robust Rectification Algorithm for Non-Calibrated Multi-view Images", in Proc. International Conference on 3D Systems and Applications (3DSA), Taipei, Taiwan, Aug. 29-30, 2018.
  • Jonghee Kim, Jinsu Kim, Seokeon Choi, Muhammad Abul Hasan, and Changick Kim, "Robust Template Matching Using Scale-Adaptive Deep Convolutional Features", Proc. IEEE Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), Kuala Lumpur, Malaysia, Dec. 12-15, 2017.
  • Seokeon Choi, Jonghee Kim, and Changick Kim, "A Selective Upscaling Method via Super-Resolution Validity Measure," Proc. IEEE International Conference on Consumer Electronics-Asia (ICCE-Asia), Seoul, Korea, Oct. 26-28, 2016.
  • 이한결, 최석언, 김창익, "좁은 기준선 영상에서 지역 기술자 성능평가", 대한전자공학회 추계학술대회, 송도 컨벤시아, Nov. 24, 2017
  • 최석언, 김종희, 김창익, "고해상도화를 위한 영역 기반 합성 가능성 예측 방법", 한국멀티미디어학회 추계학술발표대회, 경북대학교, Nov. 07, 2015.


Programming skills
  • MATLAB : Advanced level
  • C++ : Intermediate level
  • Python : Intermediate level