Sialorrhea Analysis

This repository contains a pipeline for analyzing facial features related to sialorrhea and hypomimia in Parkinson’s disease, using static video frame analysis.

Repository Structure

• find_neutral.py / find_happiness_apex.py – Identify neutral and peak-expression (apex) frames from video.
• extract_face_landmarks.py – Extract 68-point facial landmarks using dlib.
• extract_features*.py – Compute facial geometry and expression features.
• feature_analysis.ipynb – Jupyter notebook for visualizing and analyzing extracted features.
• move_booth_videos.py – Script to organize video datasets.
• significant_results*.csv – Output files containing statistical feature results.
• requirements.txt – Lists required Python packages.

Quick Start

1. Install dependencies

   pip install -r requirements.txt