FaceSync

FaceSync enables recording, audio-based synchronization, and landmark-based extraction of facial expression data from head-mounted camera videos for quantitative analysis in behavioral and social research.

Key Features:

• 3D head-mounted camera blueprints: Provides detailed 3D blueprints to construct head-mounted camera setups for facial recording experiments.
• Audio-based automated video synchronization: Automatically synchronizes multiple video recordings using audio inputs to align temporal streams.
• Manual audio alignment and inspection: Offers manual audio alignment tools and inspection workflows for reviewing and adjusting synchronization.
• Facial landmark extraction and movement plotting: Detects facial landmarks and produces graphical plots of landmark trajectories and movements over time.
• Python-based computer vision and machine learning processing: Implements processing and analysis components in Python using computer vision and machine learning algorithms for rapid extraction of expression metrics.

Scientific Applications:

• Group interaction studies: Captures naturalistic facial expressions in multi-participant interactions using head-mounted cameras.
• Behavioral and social neuroscience: Quantifies facial dynamics for experiments in social cognition, emotion, and interpersonal behavior.
• Psychology experiments: Enables objective, time-resolved measurement of facial expressions to supplement or replace manual coding.
• Video-based facial expression analysis: Supports rapid extraction of expression data from recorded video for downstream statistical or machine-learning analyses.

Methodology:

Computational methods explicitly include audio-based automated synchronization, manual audio alignment and inspection, facial landmark detection and plotting, and processing implemented in Python using computer vision and machine learning algorithms.