Profile Photo
Jisoo Kim
I put emphasis on both value creation and appropriation, taking extreme ownership,
and I am committed to developing solutions for real-world problems.

Education

Seoul National University
M.S. in Data Science (Mar 2023 – Feb 2025)
Advisor: Prof. Jaejin Lee
Sejong University
B.S. in Intelligent Mechatronics Engineering (Mar 2019 – Feb 2023)
Coursework included Programming, Artificial Intelligence, Semiconductor Process and Electrical Engineering.

Publications

[ICCV] UDC-VIT: A Real-World Video Dataset for Under-Display Cameras
Kyusu Ahn, JiSoo Kim, Sangik Lee, HyunGyu Lee, Byeonghyun Ko, Chanwoo Park, and Jaejin Lee
In Proceedings of the IEEE/CVF International Conference on Computer Vision, Honolulu, Hawai’i, USA, Oct 2025. To appear. 🏅 Highlight Paper
arXiv Project site
Integrating Spatial and Frequency Information for Under-Display Camera Image Restoration
Kyusu Ahn, Jinpyo Kim, Chanwoo Park, JiSoo Kim, and Jaejin Lee
Pattern Analysis and Applications (PAA), Springer, 2025. To appear.
arXiv Project site
[IMID] Challenges in Real-World Under-Display Camera Video Restoration
Jisoo Kim, Kyusu Ahn, and Jaejin Lee
IMID 2024: Twenty-fourth International Meeting on Information Display, Jeju, Republic of Korea, Aug 2024.

Experience

SNU Center for Optimizing Hyperscale AI Models and Platforms
Research Assistant (Mar 2025 – Present)
Current research focuses on developing efficient diffusion models for real-world, high-resolution dataset generation.
DeepLearning.AI Courses
Korean Translator
Contributed to Korean translation and subtitle synchronization for Andrew Ng and Isa Fulford’s courses, including "ChatGPT Prompt Engineering for Developers" and "Building Systems with the ChatGPT API."
CIT Code Academy
Instructor (Dec 2022 – Sep 2023)
Delivered lectures on Python and C programming, providing foundational data science education using Google Colab and Kaggle.

Projects

YOCO: You Only Cook Once
🥈 2nd Place Award, Ambient AI Competition
Seoul National University Graduate School of Data Science (Dec 2022 – Mar 2023)
YOCO is a cooking assistant AI developed with the Unbiased-Teacher v2 framework. It predicts the doneness of ingredients, making cooking easier and more convenient for users.
Project site Demo Video
E4/7
🏆 1st Place Award, 2021-2 Creative Learning Community
🥈 2nd Place Award, 10th SW·AI Hackathon
Sejong University (Sep 2021 – Jul 2022)
E4/7 is a real-time AI web platform for remote exam management. Leveraging computer vision technologies such as OCR and pose detection, the service automates the entire examination workflow—from candidate check-in to the grading and organization of handwritten answers.
Project site Demo Video1 Demo Video2
SLACK for ME
Seoul National University Graduate School of Data Science (Sep 2023 – Dec 2023)
Developed a Retrieval-Augmented Generation (RAG) based Slack chatbot that enables efficient information retrieval through both keyword and similarity-based search, without the need to store the entirety of Slack content.
Paper Slides Demo Video
Go Together: 같이가요
Seoul National University Graduate School of Data Science (Mar 2023 – Jul 2023)
A platform for quickly and easily matching event companions based on user preferences, built with Python, Django, PostgreSQL, and Neo4j.
Demo Video
4K
Sejong University (Sep 2021 – Jul 2022)
4K is an AI-based web service designed for military and law enforcement use, capable of accurately identifying specific individuals even when the lower part of their face is covered by a mask.
Demo Video

Skills

Programming: Python*, C, C++
Libraries/Frameworks: PyTorch*, TensorFlow, NumPy, Pandas, OpenCV, Matplotlib, Scikit-learn
Database: PostgreSQL, NoSQL, Neo4j
Environment: Linux*
* Primarily used

Qualifications

OPIC (Aug 2024) - IH
TOEIC (Aug 2022) - 930
Hits