Junho Cho

25.05 Updated
Read my posts
Github @junhocho
linked_in
Resume_202403.pdf

Affiliation & Education

Samsung Electronics, Sep. 2022 ~ Present
SAIT (종합기술원) → AI Center
Staff Researcher

NAVER Corp, Aug. 2018 ~ Nov 2019
Visiting Researcher

Seoul National University, Ph.D., Mar. 2015 ~ Aug. 2022
Integrated Ph.D course, Electrical and Computer Engineering

KAIST, B.S., Feb. 2010 ~ Feb. 2015
Double majored in Computer Science and Bio and Brain engineering

Experience

Samsung Electronics 산학장학생, Sep. 2021 ~ Mar. 2022

Fastcampus 오프라인 강의 강사 (딥러닝을 위한 리눅스와 GPU Camp)

SK ML Egineer과정 실습 조교, Mar, Jul, Oct. 2021, Mar, Jul. 2022

서울대학교 인공위성지구물리 연구실 자문 (위성데이터 딥러닝 프로그램 개발)

MetaMong, Oct. 2015 ~ Jan. 2016, Android App which can find your face-lookalikes, Developed face recognition and retrieval, publicized in Google HackFair 2015, Interview link

B4Sunset, Sep. 2014 ~ Dec. 2014, Blind dating service, Developed user matching algorithm

Publications (Interntional)

Junho Cho, Kyuewang Lee, Jin Young Choi, “Font Representation Learning via Paired-glyph Matching” British Machine Vision Conference (BMVC), 2022.
code

Junho Cho, Sangdoo Yun, Dongyoon Han, Byeongho Heo and Jin Young Choi, “Detecting and Removing Text in the Wild” in IEEE Access, vol. 9, pp. 123313-123323, 2021, doi: 10.1109/ACCESS.2021.3110293, September, 2021.

Jiwoong Park*, Junho Cho*, Hyung Jin Chang, Jin Young Choi, “Unsupervised Hyperbolic Representation Learning via Message Passing Auto-Encoders”, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021. (* Equally contributed)
code

Junho Cho, Sangdoo Yun, Kyoungmu Lee and Jin Young Choi, “PaletteNet: Image Recolorization with Given Color Palette” The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, NTIRE2017, July, 2017.
project

Patents (Published Applications)

Method and Apparatus with Object Detection for Autonomous Vehicle
US 2025/0131707 A1 — Lead inventor, published 24 Apr 2025
Prompt-based lightweight object-detection pipeline for autonomous vehicles.
USPTO Patent Center

Technical Skills

Python, PyTorch (ddp, mpirun), LeRobot
bash/zsh, docker, tmux, neovim, ROS, vLLM/SGLang, Flask/FastAPI
Experienced in Exynos V920, NVIDIA Jetson AGX Orin, ALOHA Table-top bimanual robot, SO-ARM100
and <english> for LLM prompting