BMI 212 (2025) Project

This is a project for BMI 212 (2025) that is to create a deep learning model on some biologically relevant dataset

Architecture

1. YouTube mp4 data for clip extraction

2. Mapped mediaPipe landmark data

3. Landmark data to CSV

4. Model testing

5. Presentation

Roles:

• Hanson - Cleaning Extracting MP4 data & Cleaning Code + EDA

• Jeff - Landmark Mapping and real-time-inference module

• Jackson - Tree-based & RNN model building

• Jayson - Logistic Regression model building