pyAprilTag

Overview

This is the python3 wrapper for AprilTag implemented in cv2cg.

Currently it is tested under Windows10(VC2017, x64)/Ubuntu 18.04, OpenCV3.4.7(official prebuilt version), and python3.7/3.6 in Anaconda3.

Version

0.0.6

Installation

Install from Pip/Conda (easiest)

0. Download file opencv_py37_conda-forge.txt to your anaconda base directory (default in windows is %USERPROFILE%)

1. try to create and activate a new conda environment with:

   conda create --name opencv_py37 python=3.7 --file opencv_py37_conda-forge.txt -c conda-forge
   conda activate opencv_py37
   

2. for Windows users:

   pip install https://github.com/ai4ce/pyAprilTag/releases/download/0.0.6/pyAprilTag-0.0.6-cp37-cp37m-win_amd64.whl
   

   for Linux users:

   pip install https://github.com/ai4ce/pyAprilTag/releases/download/0.0.6/pyAprilTag-0.0.6-cp37-cp37m-linux_x86_64.whl
   

3. Enjoy the demo:

   python -m pyAprilTag.demo
   

Install by Compiling from Source

Following is for Mac users who need to compile from source, or other users who want to try installing from source.

0. clone repository recursively, and install pre-built opencv 3.4.7 for python from conda-forge:

   git clone --recursive https://github.com/ai4ce/pyAprilTag.git
   conda install -c conda-forge opencv=3.4.7
   

1. prepare your build environment:

   #this example is for Windows built
   "C:\Users\cfeng\Anaconda3\Scripts\activate.bat"
   "C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Auxiliary\Build\vcvars64.bat"
   

2. compile the project and you are done:

   pip install -e .
   

Usage

python -m pyAprilTag.demo
# expected output:
#[ 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
# 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39]

python -m pyAprilTag.demo_plt
python -m pyAprilTag.demo_calib
python -m pyAprilTag.demo_calib_by_photo #remember to remove the calib_log folder in the current working directory
python -m pyAprilTag.vdemo_cv
python -m pyAprilTag.vdemo_plt

Contact

Chen Feng

Feel free to email any bugs or suggestions to help us improve the code. Thank you!