Copyright (C) 2014 Alina Kuznetsova This program is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 2.1 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details. You should have received a copy of the GNU General Public License along with this program. If not, see <http://www.gnu.org/licenses/>. Author: Alina Kuznetsova =================================News=========================================== - The test data collected using Intel Creative Depth Sensor can be downloaded from here: https://dl.dropboxusercontent.com/u/25197257/data.zip - The corner detector is switched to the automatic MATLAB corner detector - now you only need to select the regions where the checkerboard is and not the exact corners (-> faster) ================================================================================ Please cite the following work if you use the code: Alina Kuznetsova, Bodo Rosenhahn "On calibration of a low-cost time-of-flight camera" ECCVW 2013 @InProceedings{Kuznetsova_2014_ECCV_Workshops, author = {Alina Kuznetsova and Bodo Rosenhahn}, title = {On calibration of a low-cost time-of-flight camera}, booktitle = {The IEEE European Conference on Computer Vision (ECCV) Workshops}, month = {September}, year = {2014} } If you have any questions, please contact me: kuznetso@tnt.uni-hannover.de =========================3d-party code used====================================== The following 3d-party code is used and included with the code: 1) the ToF Calibration toolbox is based on Matlab Kinect calibration toolbox (http://www.ee.oulu.fi/~dherrera/kinect/) 2) the code uses Gaussian Kernel regression by Youngmok Yun (http://youngmok.com/gaussian-kernel-regression-for-multidimensional-feature-with-matlab-code/) 3) the code uses fast distance matrix computation by Mo Chen (http://www.mathworks.de/matlabcentral/fileexchange/24599-pairwise-distance-matrix) =========================Usage====================================== Please note, that right now the Toolbox supports only one color camera. This will be corrected in future releases. Starting new calibration, firstly call: global_vars(); %make all global variables visible in the workspace dataset_path = '%ENTER YOUR PATH HERE%'; %set the path to the folder, containing RGB, conf and depth images iterative_calib %start calibration Then follow the instructions printed in MATLAB; At some point, you will have to select checkerboard corners, but the corners might not be found correctly. If it happends, break the process (Ctrl+C) and start again; the process will be started from the same point The final calibration will be available in workspace as calib %object containing calibration parameters errors %errors at each iteration * To show color-depth overlay: function color_depth_overlay(calib,rgbfile,depthfile) %show depth-to-rgb overlay function color_depth_overlayi(calib,i) %show depth-to-rgb overlay of the i-th calibration image * To transform depth to world coordinates pn = get_dpoint_direction_Intel(p(1,:),p(2,:),calib.cK,calib.ckc); %pixel coordinates to normalized coordinates p3D = [pn.*repmat(d,2,1); d]; %coordinates in 3D (d denotes corresponding depth value) * To transform depth to rgb camera system p3Drgb = calib.cR * p3D + repmat(calib.ct,1, size(p3D,2)); * To reproject depth points to RGB camera plane function prgb = depth2rgb(p,d,calib) * Save all global variables function save_globals(fn) * Load all global variables load(filename) %standard MATLAB function * To run the calibration from the start, set the corresponding fields in calib0 to []
raminrahimi6970/ToF-Calibration
A ToF camera calibration toolbox; tested with Intel Creative; tests for Kinect 2 in progress
MATLAB