The is an updated version of the Multi Camera Self Calibration toolbox by Svoboda et al.
Links:
-
This version of the code lives on github.
-
The original website remains a good source of information. (This version of the code is being made publicly available with the permission of Tomas Svoboda.)
-
We have an online discussion group.
If you use this library, please cite:
Svoboda T, Martinec D, Pajdla T. (2005) A convenient multi-camera self-calibration for virtual environments. PRESENCE: Teleoperators and Virtual Environments. 14(4):407-422. link
At a high level, the changes are all in the usage, and not in the algorithmic domain. Those I remember are:
-
Various small changes to get to work using Octave
-
Disable plotting when running using Octave
-
Implementation of new configuration file format that lives alongside the calibration data rather than in the source code files
-
Inclusion of some sample data and tests
-
Inclusion of a python interface to run the MultiCamSelfCal application
-
Improved the interface for maximally aligning a new calibration to existing camera centers. See Align-Existing config option, and the original_cam_centers.dat input file.
The original readme file is in MultiCamSelfCal/README.txt
To test if everything is working for you, do the following:
cd MultiCamSelfCal
octave gocal.m --config=../strawlab/test-data/DATA20100906_134124/no-global-iterations.cfg
This will run most of the algorithm on some sample data. Your computer should churn for a few minutes and finally should end with some lines indicating a successful calibration (with mean reprojection error 0.62 pixels):
GNU Octave, version 3.2.3
Copyright (C) 2009 John W. Eaton and others.
This is free software; see the source code for copying conditions.
There is ABSOLUTELY NO WARRANTY; not even for MERCHANTABILITY or
FITNESS FOR A PARTICULAR PURPOSE. For details, type `warranty'.
Octave was configured for "x86_64-pc-linux-gnu".
Additional information about Octave is available at http://www.octave.org.
Please contribute if you find this software useful.
For more information, visit http://www.octave.org/help-wanted.html
Report bugs to <[email protected]> (but first, please read
http://www.octave.org/bugs.html to learn how to write a helpful report).
For information about changes from previous versions, type `news'.
arg = --config=../strawlab/test-data/DATA20100906_134124/no-global-iterations.cfg
config_dir = /home/astraw/astraw-git-root/flydra/MultiCamSelfCal/MultiCamSelfCal/../strawlab/test-data/DATA20100906_134124/
Multi-Camera Self-Calibration, Tomas Svoboda et al., 07/2003
************************************************************
Experiment name: strawlab_test
warning: The calibration file config.files.CalPmat does not exist
warning: No P mat available
warning: The calibration file config.files.CalPmat does not exist
warning: No P mat available
warning: The calibration file config.files.CalPmat does not exist
warning: No P mat available
warning: The calibration file config.files.CalPmat does not exist
warning: No P mat available
warning: No Pmat available
********** After 0 iteration *******************************************
RANSAC validation step running with tolerance threshold: 10.00 ...
RANSAC: 2 samples, 523 inliers out of 523 points
RANSAC: 1 samples, 523 inliers out of 523 points
RANSAC: 2 samples, 432 inliers out of 434 points
RANSAC: 1 samples, 362 inliers out of 362 points
522 points/frames have survived validations so far
Filling of missing points is running ...
Repr. error in proj. space (no fact./fact.) is ... 0.708677 0.688062
************************************************************
Number of detected outliers: 0
About cameras (Id, 2D reprojection error, #inliers):
CamId std mean #inliers
1 0.62 0.69 432
2 0.81 0.70 523
3 0.83 0.78 523
4 0.49 0.53 362
***************************************************************
**************************************************************
Refinement by using Bundle Adjustment
Repr. error in proj. space (no fact./fact./BA) is ... 0.714557 0.686345 0.620358
2D reprojection error
All points: mean 0.62 pixels, std is 0.61
John Stowers wrote an API to let this code be directly called from Python.
To install:
pip install multicamselfcal
To install from the git repository:
pip install -e .
To test:
pytest