Skip to content
View MinjuKim023's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report MinjuKim023

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
MinjuKim023/README.md

Minju Kim

Computer Vision · Machine Learning Research · Cognitive AI
🔬 Research Data Scientist @ Cognitive Development Lab, Indiana University Bloomington
🎓 MS Data Science, Indiana University Bloomington (2025)

Portfolio LinkedIn


Research Focus

I build computer vision pipelines to study how infants perceive and learn from their visual environment — analyzing 500,000+ egocentric head-camera images and videos collected from infants aged 4–30 months.

My work sits at the intersection of deep learning, quantitative image analysis, and developmental psychology. I translate SOTA CV/ML research into reproducible scientific tools that support academic publications and conference presentations.


🏆 Selected Work

Infant Egocentric Vision Analysis · IU Cognitive Development Lab, 2024–Present
End-to-end CV pipeline measuring visual complexity in infant environments using feature-congestion models, semantic segmentation, and multimodal video analysis. Dataset: 500K+ head-camera frames across 30 infants.

EvacSim — Multi-Agent Evacuation Simulation · GitHub · Demo
🥈 2nd Place, Luddy Hackathon 2025. Real-time evacuation tool with 9 autonomous agents (8 drones + 1 ground vehicle) built in Unity with AI NavMesh pathfinding, integrated with a React frontend via WebSocket.

Algae Classification — City of Bloomington Partnership · Report · Poster
Applied ResNet and U-Net to classify 12 algae species from microscopy images for Lake Monroe water quality monitoring. 1,287 images preprocessed and augmented to address class imbalance.

Heart Disease Risk Assessment · GitHub
ML classification models (Logistic Regression, Random Forest, XGBoost) across full, female-only, and male-only cohorts. EDA, feature engineering, and model evaluation in Python + scikit-learn.


⚙️ Technical Stack

Languages

Python R SQL MATLAB HTML

ML / Deep Learning

PyTorch TensorFlow scikit-learn OpenCV HuggingFace

Data & Visualization

NumPy Pandas Tableau Power BI

Databases & Tools

MongoDB Snowflake Git Google Cloud Linux


Always open to discussing research, collaboration, or interesting problems.

Popular repositories Loading

  1. Predicting-Telecom-Customer-Loyalty Predicting-Telecom-Customer-Loyalty Public

    Jupyter Notebook

  2. Microscopic-Algae-Classification Microscopic-Algae-Classification Public

    Python

  3. Bridging-the-Gender-Gap-in-Heart-Attack Bridging-the-Gender-Gap-in-Heart-Attack Public

    Jupyter Notebook

  4. matlabPyrTools matlabPyrTools Public

    Forked from LabForComputationalVision/matlabPyrTools

    MatLab tools for multi-scale image processing, including Laplacian pyramids, Wavelets, and Steerable Pyramids

    MATLAB

  5. Piranhas Piranhas Public

    Forked from ArturoDeza/Piranhas

    A Toolkit for creating Peripheral Architectures (NIPS 2016)

    MATLAB

  6. visual-clutter visual-clutter Public

    Forked from kargaranamir/visual-clutter

    This Python implementation of two measures of visual clutter described in the Journal of Vision 2007 in the special issue on crowding.

    Python