/dslisleedh_cv

이동헌(Dongheon Lee, dslisleedh)

Birth: 12/22/1997

Email: dslisleedh@gmail.com, dslisleedh@uos.ac.kr

Career

  • UST21 (Research & Development department)
    • AI Engineer (Mar. 2022 ~ Apr. 2023)
    • Work experiences

      • Sea-Surface-Temperature super resolution (July 2022 ~ Apr. 2023)
        SST_SR_result To upscale SST data from remote sensing, I tested two strategies: SISR and MISR. I chose NAFSR (Non-linear Activate Free Network for Super-Resolution) for SISR because of its simplicity, and TR-MISR for MISR because it is the SOTA model in the unique MISR dataset, PROBA-V. SISR (NAFSR) performed best, with a PSNR gain of +0.97 dB over MISR and +1.34 dB over bicubic interpolation at scale 4. 

      • Sea-fog generation prediction (Apr. 2022 ~ Dec. 2022)
        I tested various models to handle an extremely imbalanced time-series classification dataset. After the training is complete, I deploy the model to the actual service. Models tested:

        • Transformer
        • GFNet
        • 1D-Convolution based model

      • Anomaly detection on maritime AIS data (May 2022 ~ Nov. 2022)
        Model
        We've considered 2 ways to detect anomalies in maritime AIS data: time-series based anomaly detection and image-based anomaly detection. I tested image-based anomaly detection, which converts AIS data into track images and uses them as a feature for computational efficiency. It was hard to use existing research because our ROI covered the whole Daehan Strait and Mokpo Sea, but their ROI was only within a specific port. So I proposed a new model architecture which can recognize both track's shape and position in ROI and It showed better performance than time-series based model in our test dataset.

Papers

  • Arbitrary-Scale Downscaling of Tidal Current Data Using Implicit Continuous Representation (Arxiv, 2024)
  • PLKSR: Partial Large Kernel CNNs for Efficient Super-Resolution (Arxiv, 2024)

Education

  • Kangnam Univ. (Mar. 2016 ~ Feb. 2022)
    • B.S. degree
    • Department of Library Information Science
    • Thesis: Analysis of book characteristics that affect book lending
  • University of Seoul (Sep. 2023 ~ )
    • M.S. studuent
    • Department of Artificial Intelligence

Research Interests

  • Computer Vision
  • Image Processing
    • Super-Resolution
  • Computational Efficiency

Things I can do ...

  • Implement models in various frameworks(TF2/Keras, Pytorch, Flax(Contributor)).
  • Propose a new model architecture suitable for the situation.