AprilTags are 2D barcodes developed for robotics applications by Ed Olson. The library detects any April tags in a given image, provides the unique ID of the tag as well as its location in the image. If the camera is calibrated and the physical size of the tag known, also provides the relative transform between tag and camera.
Examples of different tag families (number of bits versus minimum Hamming distance between valid tags).
Apriltag detections in an image (see examples/apriltags_demo.cpp).
The AprilTags C++ library is available as open source under GNU LGPL version 2.1.
AprilTags mainly depends on OpenCV and Eigen3. All dependencies can be installed in Linux (tested on Ubuntu 12.04) with the following command:
sudo apt-get install subversion cmake libopencv-dev libeigen3-dev libv4l-dev
and in Mac OS X (tested on 10.8.2) with:
sudo port install pkgconfig opencv eigen3
You can check out AprilTags from our public svn repository:
svn co https://svn.csail.mit.edu/apriltags
The AprilTags library uses the pods build system in connection with cmake. Compile with
After compiling, run the example program
Ed Olson, AprilTag: A robust and flexible visual fiducial system, Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 2011