Byeongjun Kwon

My Research Topics: Depth Estimation and Dynamic 3D World Modeling

Computer Vision Researcher | Ph.D Student

Explore My Research Get In Touch
Minh-Quan

About Me

I am a Ph.D. student in Electrical Engineering at VIC Lab, Advanced Institute of Science & Technology (KAIST), under the supervision of Prof. Munchurl Kim.

My current research focuses on computer vision, depth estimation, neural rendering, and Dynamic 3D world reconstruction from images and videos.

Education

  • Ph.D. in Electrical Engineering KAIST, Feb.2024-Present
  • M.S. in Electrical Engineering KAIST, Mar.2022-Feb.2024
  • B.S. in Electrical Engineering Yonsei University, Mar.2016-Feb.2022
    (including mandatory military service, Jul. 2017 – Mar. 2019)

Research Interests

Computer Vision Depth Estimation 2D/3D Reconstruction Neural Rendering Deep Learning Computer Graphics

Technical Skills

Python Pytorch

Research Projects

One Look is Enough
View Details

One Look is Enough: Seamless Patchwise Refinement for Zero-Shot Monocular Depth Estimation on High-Resolution Images

Proposes efficient and generalizable tile-based depth refinement framework for zero-shot monocular depth estimation on high-resolution images.

ICCV 2025

Publications

Conference

One Look is Enough: Seamless Patchwise Refinement for Zero-Shot Monocular Depth Estimation on High-Resolution Images

Byeongjun Kwon, Munchurl Kim

ICCV 2025

Paper Project Page GitHub
Conference

From-Ground-To-Objects: Coarse-to-Fine Self-supervised Monocular Depth Estimation of Dynamic Objects with Ground Contact Prior

Jaeho moon, Juan Luis Gonzalez Bello, Byeongjun Kwon, Munchurl Kim

CVPR 2024

Paper Project Page GitHub
Domestic Journal

Dual-Modal Depth Estimation based on Cross Reference and Fusion for Misaligned Thermal and RGB Images

Byeongjun Kwon, Munchurl Kim

Journal of Broadcast Engineering (JBE)

PDF Cite
Challenge

Realistic bokeh effect rendering on mobile gpus, mobile ai&aim 2022 challenge: report

Andrey Ignatov, Radu Timofte, Jin Zhang (and 21 others, including Byeongjun Kwon)

ECCV 2022 Workshops

PDF

Get In Touch

Email

byeongjun.kwon@kaist.ac.kr

Laboratory

N24 Building
Room 1106
KAIST