Skip to content

ArUco markers

Detect square coded markers and recover their identities and ordered corners.

Try it on your image ↓

VISUAL WALKTHROUGHDetection
REFERENCE INPUTA code seen in perspective

The illustrative module grid is a schematic, not a valid encoded marker.

RESULTValidate and decode

A candidate must pass the dictionary or code validity checks before it yields a payload. The schematic grid above is not decoded.

STEP 03 / 03

Decode against a dictionary and reject invalid codes.

Illustrative example. The stages explain the method; they are not a live OpenCV execution.

Try it on an image

YOUR IMAGE · REAL OPENCV

Experiment at pixel level

Open full lab ↗

Detect ArUco markers in the selected dictionary and draw their corners and IDs. Upload a marker photo to get detections.

The engine loads on your first run. Your images stay in this browser.

Input448 × 320
OutputWaiting for a result

Scroll over either image to zoom at the pointer. Use the scrollbars to pan both views over the same relative area. Zoom is relative to the input; pixel coordinates belong to each image. Warps can change scene correspondence.

Pixel inspector RGBA · native values · matched scale · 9 × 9 output pixels
Hover to inspect. Click to pin a pixel.
Input
Select a pixel

Output
Select a pixel

Sample models and licenses

When to use it

Build fiducial tracking, calibration boards and known-geometry pose inputs.

How it works

  1. 01Locate square-like candidates and refine their corners.
  2. 02Rectify each candidate into its code grid.
  3. 03Decode against a dictionary and reject invalid codes.

square candidate → rectified bit grid → dictionary ID

What to tune

Use the same dictionary when generating and detecting. Supply the marker’s physical size and camera calibration for pose estimation.

Where it breaks down

Detection returns image geometry, not a fully calibrated world pose. Motion blur, glare and too-small markers reduce reliability.

TypeScript API

Open an entry for its exact overloads, parameter descriptions, result ownership and pinned upstream source.

All of these calls execute on the CPU. Native objects need explicit disposal. See matrices and ownership and build compatibility.