Jae-Hyeok Lee

Ph.D. Candidate @ KAIST

my_pic_fotor.png

I am currently a Ph.D. candidate at KAIST, advised by professor Dae-Shik Kim. I have primarily researched computer vision in various application domains, including medical, satellite, and facial imaging fields.

Initially, I was drawn to practical tools and methods that could benefit people. However, over time, my focus shifted to innovations that not only bring joy to people but also have the potential to fundamentally transform their lives.

My recent research interests lie in 3D vision and computer graphics, including the Neural Radiance Field (NeRF). I believe that emerging technologies like NeRF will fundamentally change the way people capture and store time and space.

I am actively conducting research on pivotal technologies for this future, with a keen interest in the editing of 3D spaces and content generation within them. I am particularly interested in topics such as 3D editing, 3D content generation, neural radiance field, neural representation, and their various applications.

Selected Publications

2023

  1. ice-nerf.gif
    ICE-NeRF: Interactive Color Editing of NeRFs via Decomposition-Aware Weight Optimization
    Jae-Hyeok Lee, and Dae-Shik Kim
    International Conference on Computer Vision (ICCV), 2023

2020

  1. mcsip.gif
    MCSIP Net: Multichannel Satellite Image Prediction via Deep Neural Network
    Jae-Hyeok Lee, S. Sangmin Lee, Hak Gu Kim, Sa-Kwang Song, and 2 more authors
    IEEE Transactions on Geoscience and Remote Sensing (TGRS), 2020

2019

  1. probenet.png
    Probenet: Probing Deep Networks
    Jae-Hyeok Lee, Seong Tae Kim, and Yong Man Ro
    International Conference on Image Processing (ICIP) (Oral), 2019