PupilSense: A Novel Application for Webcam-Based Pupil Diameter Estimation

2024-07-15

AI Summary

This study addresses the challenge of accurately measuring pupil diameter in everyday settings without specialized hardware. We propose a lightweight, end-to-end CNN model that performs real-time regression estimation of left and right pupil diameters—and quantifies eye aspect ratio during blinks—using only standard webcam video streams. The method incorporates adaptive eye-region cropping and normalization preprocessing, augmented by Grad-CAM to generate physiologically interpretable heatmaps. Key contributions include: (1) the first open-source webcam-based pupil annotation dataset; (2) a high-accuracy estimation algorithm achieving a mean absolute error of ±0.12 mm; and (3) cross-device real-time deployability alongside a comprehensive visualization and analysis framework. Both code and dataset are publicly released, establishing a low-cost, interpretable foundation for behavioral science research and remote health monitoring.

Technology Category

• Computer Vision: Motion & Tracking
• Machine Learning: Calibration & Uncertainty Quantification
• Application Domains: Humanities & Computational Social Science

Application Category

• Security and Privacy: Large-scale security measurements
• Web Mining and Content Analysis: Large pretrained models with web data
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Measuring pupil diameter is vital for gaining insights into physiological and psychological states - traditionally captured by expensive, specialized equipment like Tobii eye-trackers and Pupillabs glasses. This paper presents a novel application that enables pupil diameter estimation using standard webcams, making the process accessible in everyday environments without specialized equipment. Our app estimates pupil diameters from videos and offers detailed analysis, including class activation maps, graphs of predicted left and right pupil diameters, and eye aspect ratios during blinks. This tool expands the accessibility of pupil diameter measurement, particularly in everyday settings, benefiting fields like human behavior research and healthcare. Additionally, we present a new open source dataset for pupil diameter estimation using webcam images containing cropped eye images and corresponding pupil diameter measurements.

Problem

• Enables pupil diameter estimation using standard webcams
• Makes measurement accessible without specialized equipment
• Expands accessibility for human behavior research and healthcare

Innovation

• Webcam-based pupil diameter estimation
• Open source dataset for pupil measurement
• Detailed analysis with class activation maps