MMART is a command line tool to record raw data from several sensors that are connected to a computer. Prior to usage MMART has to be configured via an XML-file. To start a recording simply run recorder.exe, e.g. by using a command line shell of your choice. For setting up the system you can either use a pre-built package (TODO) or build it yourself from source.
- Intel RealSense
- Thalmic Labs Myo
- SoftKinetic DepthSense
...
Several options can be set via the XML configuration file:
- Devices that shall be used
- Streams of that devices to record
- Target path
- Install Visual C++ Redistributable Packages für Visual Studio 2013
- Install RealSense SDK Runtime and F200 DCM (v1.4)
- PCL 1.7.2 from http://unanancyowen.com/?p=1255&lang=en
- HDF5-1.8.15
- Install Myo Connect
-
Boost 1.55: sudo apt-get install libboost1.55-all-dev
-
Eigen 3: sudo apt-get install libeigen3-dev
Set environment variable: export EIGEN3_INCLUDE_PATH=/usr/include/eigen3/
- Google Glog: sudo apt-get install libgoogle-glog-dev
Set environment variable: export GLOG_ROOT=/usr/
-
Google Protobuf: sudo apt-get install libprotobuf-dev sudo apt-get install protobuf-compiler
-
HDF 5: sudo apt-get install libhdf5-dev
-
OpenCV: sudo apt-get install libopencv-dev
-
OpenGL: sudo apt-get install mesa-common-dev sudo apt-get install freeglut3-dev
-
QT 5: URL: http://download.qt.io/official_releases/online_installers/qt-unified-linux-x64-online.run
Set environment variable: export Qt5Widgets_DIR=[PATH_TO_QT]/Qt/5.2.1/gcc_64/
- Leap SDK (2.0.4) URL: https://developer.leapmotion.com/downloads/skeletal-beta?platform=linux&version=2.0.4.17546
Set environment variable: LeapSDK_DIR=PATH_TO_LEAP_SDK
sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libboost-all-dev libhdf5-serial-dev
cmake -G "GENERATER" -DCMAKE_BUILD_TYPE=Debug -DWITH_EYE-TRACKING=ON -DWITH_GESTURE_RECOGNITION=ON PATH_TO_AV_NUI
cmake -G "KDevelop3 - Unix Makefiles" -DCMAKE_BUILD_TYPE=Debug -DWITH_EYE-TRACKING=ON -DWITH_GESTURE_RECOGNITION=ON -DQt5Widgets_DIR=~/Qt/5.2.1/gcc_64/lib/cmake/Qt5Widgets -DGLOG_ROOT=/usr/ -DEIGEN3_INCLUDE_PATH=/usr/include/eigen3/ -DLeapSDK_DIR=/home/m_weber/develop/av-nui/3rdparty/x64/LeapSDK-2.1.0/ ~/develop/av-nui
cmake -G "Eclipse CDT4 - Unix Makefiles" -DCMAKE_BUILD_TYPE=Debug -DWITH_EYE-TRACKING=ON -DWITH_GESTURE_RECOGNITION=ON -DQt5Widgets_DIR=~/Qt/5.3/gcc_64/lib/cmake/Qt5Widgets -DGLOG_ROOT=/usr/ -DEIGEN3_INCLUDE_PATH=/usr/include/eigen3/ -DLeapSDK_DIR=/home/m_weber/develop/av-nui/3rdparty/x64/LeapSDK-2.2.5/ -DPMDSDK_DIR=/home/m_weber/develop/av-nui/3rdparty/x64/PMDSDK ~/develop/av-nui
- cmake
- Visual Studio 2013 Community
Always choose x64 version of a dependency!
- HDF5-1.8.15
- PCL 1.7.2 from http://unanancyowen.com/?p=1255&lang=en
- SoftKinetic SDK 1.4.5 from web page (registration required)
...
could be automated by script
Please cite the following paper (BibTeX) if you use MMART:
@inproceedings{Barz:2016:MMA:2968219.2971459,
author = {Barz, Michael and Moniri, Mohammad Mehdi and Weber, Markus and Sonntag, Daniel},
title = {Multimodal Multisensor Activity Annotation Tool},
booktitle = {Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing: Adjunct},
series = {UbiComp '16},
year = {2016},
isbn = {978-1-4503-4462-3},
location = {Heidelberg, Germany},
pages = {17--20},
numpages = {4},
url = {http://doi.acm.org/10.1145/2968219.2971459},
doi = {10.1145/2968219.2971459},
acmid = {2971459},
publisher = {ACM},
address = {New York, NY, USA},
keywords = {data annotation, data capture, multimodal, multisensor},
}