CamDistance - Automatic Camera Distance Detection for Research
Overview
CamDistance is a web-based tool for automatic camera distance detection using facial landmark detection. Developed specifically for experimental psychology research, this tool automatically measures the viewing distance between participants and their computer screens using real-time facial recognition technology.
High Precision
Achieves ±2-3 cm precision for distances between 40-100 cm, making it ideal for controlled experimental conditions.
Key Features
Automatic Measurement
Uses MediaPipe Face Mesh for real-time facial landmark detection and distance calculation.
High Accuracy
Achieves ±2-3 cm precision for distances between 40-100 cm using advanced algorithms.
Web-Based
No software installation required, works in any modern browser with camera access.
Framework Integration
Built as a jsPsych plugin with JATOS compatibility for seamless research integration.
Real-Time Feedback
Provides live measurement feedback to participants during experimental sessions.
Stabilization
Temporal smoothing and outlier rejection algorithms ensure reliable measurements.
How It Works
- Face Detection: Uses MediaPipe to detect 468 facial landmarks in real-time
- Interpupillary Distance (IPD) Measurement: Calculates pixel distance between eye landmarks
- Geometric Calculation: Applies pinhole camera model: Distance = (Real_IPD × Focal_Length) / Pixel_IPD
- Stabilization: Averages measurements over time and filters outliers
- Result Display: Shows measured distance in centimeters with real-time updates
Research Applications
In psychological experiments, particularly those involving visual stimuli, the distance between the participant and the screen critically affects:
- Visual Angle: The apparent size of stimuli on the retina
- Stimulus Perception: How participants perceive and interact with visual elements
- Data Validity: Ensuring consistent experimental conditions across participants
- Remote Studies: Enabling precise distance measurement in online experiments
Traditional methods rely on manual measurement with rulers, which is time-consuming, error-prone, and impossible in remote experiments. CamDistance automates this process using computer vision technology, making it perfect for both laboratory and online research.