README

AUTHOR
======
  Gaurav Pandey.
  email: pgaurav@umich.edu

  Redistribution and use in source and binary forms, with
  or without modification, are permitted under the GNU
  Lesses General Public License (see LICENSE)

MODIFIED BY
===========
(2019)
  David Wong
  email: david@lifecyclist.co.nz
(2017)
  Andreas L. Flaten
  email: andreas.flaten@ntnu.no

ABOUT
=====
  This is a reference implementation of mutual information
  (MI) based algorithm for automatic extrinsic calibration
  of a 3D laser scanner and optical camera system. You can
  find specific details in the original paper:

  "Automatic Targetless Extrinsic Calibration of a 3D Lidar
  and Camera by Maximizing Mutual Information." in AAAI,
  special track on robotics, July 2012, Toronto, Canada.

LIBRARIES USED
==============
  This version is implemented with the help of some external
  code:
  1) GSL          : http://www.gnu.org/software/gsl/
  2) OpenCV 3     : http://opencv.willowgarage.com/wiki/
                    (Lower versions of opencv might not work)
  3) Glib         : http://developer.gnome.org/glib/stable/
  4) CMake        : http://www.cmake.org/

COMPILATION INSTRUCTIONS
========================
  1. Go to build/ and run "cmake ../". It will generate the
     required make files.
  2. Run "make". It will generate the binary "mi-extrinsic-calib"
     in the directory bin/

USAGE INSTRUCTIONS
==================
  1. Go to config/ and update the default config file (if needed).
     The config file contains all the inputs required by the algorithm.
     This config file contains intrinsic parameters of a laser scanner
     and an omnidirectional camera used for developing this algorithm.
     Sample data from this laser scanner and the omnidirectional camera
     is provided in bin/data/.
  2. Go to bin/ and run "./mi-extrinsic-calib". If you do not
     provide any argument then it will use the default config
     file "config/master.cfg" that corresponds to a Ladybug3 omni-
     directional camera. A sample config file for a monocular camera
     setup is given as "config/monocular-cam.cfg"
  3. At the end of the program two output files are created in bin/.
     "calib_param.txt" stores the estimated calibration parameters.
     "calib_cov.txt" stores the estimated 6x6 covariance matrix.