Overview
The eyefeatures library provides tools for extracting, analyzing, and visualizing eye-tracking data.
It follows scikit-learn’s API design with fit/transform methods, making it compatible
with scikit-learn pipelines.
Key Features
Extraction of common eye-tracking features (fixations, saccades, regressions).
Blinks detection from pupil signal.
Statistical analysis of eye movement patterns and direct usage for ML tasks.
Algorithms like Markov Transition Field, Hilbert Curve calculation, Vietoris-Rips filtration for complex analysis and potential usage in Deep Learning architectures.
Visualization tools for exploring gaze/fixations patterns.
Benchmark data loading utilities (Parquet datasets, column conventions for keys/labels/meta).
scikit-learncompatible transformers forPipelineintegration.