ArUco Marker Distance Measurement
A computer-vision experiment measuring distance to ArUco markers with camera calibration and pose estimation.
Problem
Estimate the distance from a camera to an ArUco marker in a live video feed using calibration data.
Context
An exploratory computer-vision project published as open source under the MIT license; no production deployment or validated accuracy claim.
Constraints
- A one-time calibration step with the checkerboard images is required before any measurement.
- Distance estimates are only valid for the specific marker the pose estimation is configured for.
Decisions
- Calibrate the camera once with a custom checkerboard pattern and persist the camera matrix and distortion coefficients, rather than recalibrating on every run.
- Use OpenCV's ArUco detection and pose estimation to compute the camera-to-marker distance from the calibrated parameters.
Outcomes
- Working demonstration video in the repository showing a live distance readout; no accuracy figures are claimed.
The repository has two parts: a calibration script that processes checkerboard images with OpenCV’s camera calibration and saves the parameters, and a main script that loads them, detects the marker in the camera feed, estimates its pose, and displays the computed distance. It is an experiment built with Python, OpenCV, and NumPy - an exploration of the technique, not a measurement product.