└── README.md


/README.md:
--------------------------------------------------------------------------------
 1 | # A list of tools for multi-camera calibration
 2 | 
 3 | ## CamOdoCal
 4 | <br>
 5 | The package comes with both intrinsic and extrinsic calibration for multi-camera system. The advantage of this technique is that it does not require overlapping fields of view between cameras and thus this is applicable for the rear camera. A plus advantage is that this work has also been tested on a van mounted with a four-camera rigs on the roof, which is similar to our system. In order to determine the scale metric between cameras, it requires odometry data which can be from the wheel or GPS/IMU system. The package is developed in C++ on Ubuntu platform.<br>
 6 | 
 7 | __Tool:__ [Github](https://github.com/hengli/camodocal)<br>
 8 | __Platform:__ C++ executable on Ubuntu<br>
 9 | __Referrence:__ Bo, L., Heng, L., Koser, K., Pollefeys, M., 2013. A multiple-camera system calibration toolbox using a feature descriptor-based calibration pattern.<br>
10 | 
11 | 
12 | ## BACS - Bundle Adjustment for Camera System
13 | <br>
14 | This package implements a bundle adjustment algorithm for ominidirectional and multi-view cameras, which can be applied to obtain the pose of the cameras in the rig.<br>
15 | 
16 | __Tool:__ [Project Page and Code](http://www.ipb.uni-bonn.de/data-software/bacs/)<br>
17 | __Platform:__ MATLAB<br>
18 | __Referrence:__ Schneider, J., Schindler, F., Labe, T., Forstner, W., 2012. Bundle adjustment for multi-camera system with points at infinity.<br>
19 | 
20 | 
21 | ## AMCC Toolbox - Automatic Multi-Camera Calibration
22 | <br>
23 | This toolbox uses a chessboard as a calibration object. This technique requires a high degree of overlap between cameras. The toolbox offers both intrinsics and extrinsics calibration. Another advantage point of this tool is that it offers parallel processing so speed up the computation.<br>
24 | 
25 | __Tool:__ [Project Page](https://bitbucket.org/michaeldwarren/amcctoolbox/wiki/Home)<br>
26 | __Platform:__ MATLAB<br>
27 | __Code:__ `git clone https://bitbucket.org/michaeldwarren/amcctoolbox.git`<br>
28 | __Referrence:__ Warren, M., McKinnon, D., Upcroft, B., 2013. Online calibration of stereo rigs for long-term autonomy. <br>
29 | 
30 | 
31 | ## MultiCamSelfCal - Multi Camera Calibration Toolbox
32 | <br>
33 | This tool is implemented in MATLAB offering a complete and fully automatic calibration of multi-camera setups. Note that the minimum number of cameras in the rig is 3 cameras. __This is important: this technique requires a calibration to be done in a dark room as the calibration object is a laser pointer.__ In this case, it might be difficult for a vehicle equipped with cameras to be fit in a dark room that is large enough. Another concern point is that their system has multiple cameras pointing inward like a teleconference system so it might not work for a system such as cameras mounted on a vehicle. <br>
34 | 
35 | __Tool:__ [Project Page and Code](http://cmp.felk.cvut.cz/~svoboda/SelfCal/)<br>
36 | __Platform:__ MATLAB and possibly Octave compatible<br>
37 | __Referrence:__ Svoboda, T., Martinec, D., Pajdla, T., 2005. A convenient multi-camera self-calibration for virtual environments. <br>
38 | 
39 | 


--------------------------------------------------------------------------------