Yejin Lee 👩‍💻

Google Scholar ⋄ Github

📧 Email: [email protected]

---

Education 🎓

Hallym University (Kangwon, South Korea)
B.S. in Computer Science | Feb. 2023
GPA: 3.8/4.5 (Major)
Related Courses: Deep Learning, Artificial Intelligence, Computer Vision, NLP, Linear Algebra, etc.
Graduate Curriculum: Advanced Topics in Image Processing, Special Lecture on Machine Learning

---

Research Interests 🔍

• Deep Learning
• Computer Vision
• Image Processing

---

Research Experience 🧑‍🔬

Research on Developing an AI Model for Camera Calibration for Indoor CCTV Video Pedestrian Re-identification
Apr. 2023 - Dec. 2023
Supervisor: Dr. Haesol Park, Korea Institute of Science and Technology (KIST)

• Developed a feature using the mathematical model of radial distortion lenses to implement distortion.
• Inferring Camera Parameters and Distortion Coefficients from a Single Image Implementing the distortion function.

A Deep Learning-Based Coyote Detection System Using Audio Data
Aug. 2022 - Dec. 2022
Supervisors: Anthony H. Smith and Yaqin Mia Wang, Purdue University (sponsored by IITP)

• Implementation of Baseline Audio/Image Learning Models.
• Implementation of Audio Feature Extraction Code.
• Extraction and Visualization of Audio Features for Each Class to Construct a Dataset for Image Learning.

Survey Adversarial Attacks and Neural Rendering
Aug. 2022 - Oct. 2022
Supervisor: Prof. Jonguk Hou, Hallym University

• Comparative Analysis of Related Research.
• Investigation of Neural Rendering and 3D Representation Technologies.
• Analysis of Adversarial Attacks on 3D Data and Neural Rendering.

Deep Learning-Based Rapid On-Site Evaluation and Prognosis Predicting Model Using 3D Holographic Microscopy and Single Cell RNA Sequencing in Pancreatic Cancer
Jun. 2022 - Oct. 2022
Supervisor: Prof. Jonguk Hou, Hallym University

• Removed missing values from the dataset containing metadata.
• Implemented 4-fold cross-validation to prevent overfitting on the test set.
• Improved baseline accuracy from 0.93 to 0.96.

---

Publications 📚

1. H. Jung, B. Kwon, Y. Kim, et al. "A deep learning-based coyote detection system using audio data," in 2023 International Conference on Artificial Intelligence in Information and Communication (ICAIIC), 2023.

2. Lee, Yejin, B. S. Shim, and J.-U. Hou "Survey adversarial attacks and neural rendering," 2022.

---

Development Experience 💻

Medical Image AI Challenge 2023: Pathology Data
Dec. 2023 - Jan. 2024
Team Member: Seoul National University Hospital

• Predicted melanoma recurrence by utilizing pathological images and clinical data.
• Data Pre-processing for Whole-Slide Image Preprocessing.
• Ranked 6th out of 62 teams.

Future Technology Challenge 2023
Jun. 2023 - Aug. 2023
Team Member: CJ Logistics

• Developed an algorithm to recognize the quantity and type of individual products in a cart image after training on individual product images.
• Generated Synthetic Dataset for Multi-Object Detection.
• Ranked 4th out of 90 teams.

SK telecom FLY AI Challenger
Aug. 2022 - Oct. 2022
Team Member: SK telecom

• Development of a face recognition model.
• Implementation of an API for serving the ChatGPT model.
• Prompt engineering for generative models.
• Ranked 1st out of 10 teams.

---

Achievements 🏆

• 1st Place, SKT Fly AI Hackathon, awarded by SK telecom, 2023
• 1st Place, SW Capstone Design Competition, awarded by Hallym University, 2022
• 1st Place, Healthcare BigData Hackathon, awarded by Hallym University, 2021

---

Skills 💡

• Programming Languages: Python, C/C++, Java
• Machine Learning Tools: Pytorch, TensorFlow
• Environment: Linux
• Tooling: Docker, Github, Gitlab
• Language: TOEIC 815, OPIC IM2

---

Activities 🌍

K-Mate Program
Aug. 2023 - Dec. 2023
Team Reader, Korea Institute of Science and Technology (KIST)

• Cultural Exchange Program with International Colleagues.

Connective AI Workshop 2023
Aug. 2023 - Dec. 2023
Staff, Korea Institute of Science and Technology (KIST)

• Academic Conference Management and Support.

---