ArUco Tracker

Marker-based head tracking using printed ArUco patterns

The ArUco tracker uses printed fiducial markers (square patterns with unique IDs) to track head position. It’s one of the easiest trackers to set up since it requires no special hardware - just a webcam and a printed marker.

How It Works

ArUco markers are square patterns with a black border and a unique binary pattern inside. The tracker detects these markers in the camera image and calculates the 6DOF (position and orientation) pose based on the marker’s size and appearance.

ArUco tracking is based on the ArUco library from OpenCV, which provides robust marker detection even in challenging lighting conditions.

Requirements

Hardware

• Webcam: Any standard webcam (640x480 or higher recommended)
• Printed marker: ArUco marker printed on paper or cardboard
• Mounting: Way to attach marker to head (hat, headband, cap)

Printing the Marker

You can generate ArUco markers using:

• Online generators: Search for “ArUco marker generator”
• OpenCV’s ArUco module
• The marker should be printed at actual size (typically 5-10cm square)

Print the marker with high quality on stiff paper or cardboard. Ensure the marker stays flat when mounted - warping will affect tracking accuracy.

Setup Instructions

1. Print and Mount Marker

   1. Print an ArUco marker (recommend 7x7cm or larger)
   2. Mount on stiff backing if using regular paper
   3. Attach to hat, headband, or cap brim
   4. Ensure marker faces the camera and stays flat

   Larger markers are detected more reliably at greater distances. Use at least 5x5cm for desktop use.

2. Select Camera

   In OpenTrack tracker settings:

   Camera: Select your webcam
   Resolution: 640x480 (or higher for better accuracy)
   Force FPS: Default (or 30/60 for consistency)
   Use MJPEG: Enable if supported
   
3. Configure Field of View

   Set the camera’s field of view to match your webcam:

   Field of View: 56 degrees (typical webcam)
   

   If unknown, you can measure it:

   1. Place an object of known width at a known distance
   2. Measure when object fills frame width
   3. Calculate: FOV = 2 * arctan(object_width / (2 * distance))
4. Set Head Position Offset

   Configure the offset from the marker to your head’s rotation center:

   Head Position X: 0mm (left/right from marker center)
   Head Position Y: 0mm (up/down from marker center)  
   Head Position Z: 0mm (forward/back from marker center)
   

   You can also use the calibration tool:

   1. Click Calibrate in tracker settings
   2. Rotate your head while keeping shoulders still
   3. The offset will be calculated automatically
5. Test Tracking

   1. Start tracking in OpenTrack
   2. View the video preview
   3. The detected marker should be highlighted with a green outline
   4. Check that head movements are tracked smoothly

   If marker is not detected:

   • Ensure good lighting (avoid shadows on marker)
   • Check marker is flat and not warped
   • Move closer to camera
   • Try different camera angle

Configuration Options

Camera Settings

Option	Default	Description
Camera Name	-	Select your webcam
Force Resolution	0 (auto)	Set specific resolution (640x480, 1280x720, etc.)
Field of View	56°	Camera horizontal FOV in degrees
Force FPS	Default	Lock framerate (30, 60, 120, etc.)
Use MJPEG	false	Enable MJPEG compression

Available Resolutions

• 320x240 - Low quality, high FPS
• 640x480 - Standard, good balance
• 1280x720 - HD, better accuracy
• 1920x1080 - Full HD, best accuracy

Higher resolutions increase accuracy but may reduce framerate.

Head Position Offset

Option	Default	Description
Head Position X	0	Left/right offset in millimeters
Head Position Y	0	Up/down offset in millimeters
Head Position Z	0	Forward/back offset in millimeters

Advanced Features

Adaptive Threshold

The ArUco tracker uses adaptive thresholding to handle varying lighting conditions. The algorithm automatically adjusts detection parameters based on the scene.

Region of Interest (ROI)

Once the marker is detected, the tracker uses a region of interest search to improve performance:

• Reduces computational load
• Increases framerate
• Makes detection more robust
• Automatically resets to full-frame search if marker is lost

Automatic Recovery

If the marker is lost for more than 0.35 seconds, the tracker automatically:

1. Resets to full-frame detection
2. Cycles through different detection parameters
3. Adjusts threshold adaptively
4. Returns to ROI mode once marker is reacquired

Troubleshooting

Marker not detected

Lighting issues:

• Ensure even lighting on marker
• Avoid shadows or reflections on marker
• Don’t place marker in direct sunlight
• Avoid low-light conditions

Marker quality:

• Reprint marker with higher quality printer
• Ensure marker is perfectly flat
• Check for smudges or damage on marker
• Use stiff backing to prevent warping

Distance:

• Move closer to camera
• Use larger marker
• Increase camera resolution

Jittery or unstable tracking

• Use larger marker for more stable detection
• Increase camera resolution
• Improve lighting conditions
• Use higher framerate camera
• Enable filtering in OpenTrack filter settings
• Ensure marker is rigidly mounted

Tracking loses marker easily

• Increase marker size
• Improve contrast (check printer settings)
• Mount marker more perpendicular to camera
• Reduce head rotation range
• Check for camera auto-exposure issues

Wrong pose orientation

• Verify head position offset is correct
• Re-run calibration procedure
• Check marker is mounted right-side up
• Ensure marker isn’t mirrored or flipped

Low framerate

• Reduce camera resolution
• Enable MJPEG compression
• Close unnecessary applications
• Use USB 3.0 port for camera
• Disable video preview when not needed

Advantages and Limitations

Advantages

• No special hardware required
• Easy to set up
• Works with any webcam
• Reliable marker detection
• Low cost (just print a marker)
• Good accuracy for gaming

Limitations

• Marker must be visible to camera
• Limited head rotation range (~±60°)
• Marker can be uncomfortable to wear
• Affected by lighting conditions
• Less accurate than IR point tracking
• Paper markers can warp over time

Tips for Best Results

1. Print Quality: Use a laser printer for crisp, high-contrast markers
2. Marker Size: Larger markers (8-10cm) work better for desktop distances
3. Mounting: Mount marker on cap brim or headband, keep it flat and rigid
4. Lighting: Use diffuse, even lighting - avoid harsh shadows
5. Camera Position: Mount camera at eye level, about 50-80cm away
6. Background: Keep background uncluttered to help detection algorithm

Comparison with Other Trackers

Feature	ArUco	PointTracker	NeuralNet
Setup difficulty	Easy	Medium	Very Easy
Hardware cost	Very Low	Medium	Very Low
Accuracy	Good	Excellent	Good
Latency	Low	Very Low	Medium
Marker required	Yes	Yes (IR LEDs)	No
Rotation range	±60°	±90°	±70°

See Also

• PointTracker - Higher precision IR LED tracking
• NeuralNet Tracker - No marker required, AI-based
• Hardware Guide - General hardware setup information