/3DTK

!!! TEST ONLY !!! DO NOT USE !!!

Primary LanguageRoff

3DTK - The 3D Toolkit

Build Status Build status

About

The 3D Toolkit provides algorithms and methods to process 3D point clouds. It includes automatic high-accurate registration (6D simultaneous localization and mapping, 6D SLAM) and other tools, e.g., a fast 3D viewer, plane extraction software, etc. Several file formats for the point clouds are natively supported, new formats can be implemented easily. Check out our website (threedtk.de) for more information.

Build Instructions

For installation instructions on Ubuntu, Debian and macOS, please have a look at the INSTALL file.

To compile the project simply call make. This will configure slam6d using the default settings. If you wish to configure the project using custom settings do: make config. This command requires ccmake be installed on your system. Alternatively you may change into the build directory .build and configure the project with your preferred cmake configurator, i.e.:

cd .build && cmake -i ../

For Microsoft Windows, use the cmake-gui application provided by cmake to configure and generate project files for the appropriate version of Microsoft Visual Studio C++ of your system. Use the INSTALL target to built the entire project. Executables (and .dll's) will then reside in the windows folder. For running the binaries you need to install the proper redistributable package. Alternatively on Windows, execute the batch script build.cmd either from the command line or from the Explorer. It will download and extract the right versions of boost and opencv and then run cmake with the right configuration options.

Some Boost libraries (graph, regex, serialization, filesystem, interprocess) are needed to compile the slam6D program. 3DTK is not compatible with boost version 1.49. Please make sure to use a different version.

3DConnexion Space Mouse Support

The Applications show and qtshow supports navigation with the 3DConnexion SpaceMouse. The left key on the SpaceMouse Wireless/Compact fixes the rotation and the right key fixes the translation. The keyset for other models can be different.

Linux

No further steps are required for the basic functions. For full support the libspnav and the spacenavd daemon (show on GitHub) is required. Then build 3DTK with the WITH_3DMOUSE option.

macOS

It's required to install the official driver and build 3DTK with the WITH_3DMOUSE option.

Getting Started

For a detailed explanation of the programm, its usage, etc., please refer to the high level documentation (esp. sections 4-6, for starters) and here for creating videos with 3DTK. Further a reference manual can be found in doc/html/index.html after building the doxygen domcumentation (type in make docu to get the HTML files).

Important:

Take care to register scans first (bin/slam6D) before trying to display them (bin/show), and think about using the point reduction (see section 6) for a much faster scan matching result. Extremely large scans might need to be reduced (using bin/scan_red) before registration. This will write reduced scans in the uos format into a directory reduced in the data directory.

Datasets:

Three example scans are included in the dat directory, several larger data sets can be downloaded from the data repository.

Examples:

slam6d and show

Using the data set in the slam6d repository:
bin/slam6D -m 500 -R 5 -d 25.0 --metascan dat
bin/show dat
bin/slam6D --max=500 -r 10.2 -i 20 --metascan dat
bin/show dat
Using hannover1.tgz from data repository:
bin/slam6D -s 1 -e 65 -r 10 -i 100 -d 75 -D 250 --epsICP=0.00001
           -I 50 --cldist=750 -L 0 -G 1 /YOUR/DOWNLOAD/DIRECTORY/dat_hannover1
bin/show -s 1 -e 65 /YOUR/DOWNLOAD/DIRECTORY/dat_hannover1
Using hannover2.tgz from data repository:
bin/slam6D -q -r 10 -f rts -s 23 -d 75 -L 4 --cldist=1500 -G 1 -D -1
           --DlastSLAM 250 --graphDist 200 -I 50 /YOUR/DOWNLOAD/DIRECTORY/hannover2
bin/show -f rts -s 23 /YOUR/DOWNLOAD/DIRECTORY/hannover2
Using kvarntorp_mine.tgz (dat_mine1) from data repository:
bin/slam6D -s 1 -e 76 -r 10 -m 3000 -d 50 -i 1000 --epsICP=0.000001
           -I 50 -D 75 --clpairs=5000 -f old /YOUR/DOWNLOAD/DIRECTORY/dat_mine1/
bin/show -s 1 -e 76 -m 3000 -f old /YOUR/DOWNLOAD/DIRECTORY/dat_mine1/
Using bremen_city.zip from data repository:
bin/scan_red -s 0 -e 12 -r OCTREE -v 10 --octree 0 -f RIEGL_TXT /YOUR/DOWNLOAD/DIRECTORY/bremen_city
bin/slam6D -a 2 -q /YOUR/DOWNLOAD/DIRECTORY/bremen_city/reduced -f uos -d 150
            -s 0 -e 12 --anim=1 -n /YOUR/DOWNLOAD/DIRECTORY/bremen.net
		  -G 1 -D 100 -i 0 -I 50 -p --epsSLAM=0.0
bin/show -s 0 -e 12 /YOUR/DOWNLOAD/DIRECTORY/bremen_city/reduced
Using UniKoblenz_CampusTour3_OsnabrueckFormat.tar.gz from data repository:
bin/slam6D -s 1 -e 320 -r 20 -i 300 --epsICP=0.000001 -d 45 -D 45
           -f uos --algo=2 -l 10 -L 4 -I 100 -G 1
           /YOUR/DOWNLOAD/DIRECTORY/UniKoblenz_CampusTour3_OsnabrueckFormat/
bin/show -s 1 -e 320 -f uos /YOUR/DOWNLOAD/DIRECTORY/UniKoblenz_CampusTour3_OsnabrueckFormat/

Detecting Planes

For detecting planes compile with the WITH_PLANE option. Adapt the settings in bin/hough.cfg for your data set. Using the data set in the slam6d repository, no modification of bin/hough.cfg necessary:

bin/planes -s 0 dat
bin/show -s 0 -e 0 dat -l dat/planes/planes.list
Using bremen_city.zip from data repository:

Adapt these settings in bin/hough.cfg:

RhoNum              500
RhoMax              5000
MaxPointPlaneDist   50.0
MinPlaneSize        50
PointDist           100.0
bin/planes -f riegl_txt -s 0 /YOUR/DOWNLOAD/DIRECTORY/bremen_city/ -r 50 -O 1 -m 5000
bin/show -s 0 -e 0 /YOUR/DOWNLOAD/DIRECTORY/bremen_city/ -f riegl_txt -l dat/planes/planes.list -r 10 -O 1 -m 5000

Frequent Use

If you find yourself using 3DTK frequently, read about Configuration Files to find out how not to have to type as much on the command line.