Skip to content

Unix Installation

Jump to bottom
Tadas Baltrusaitis edited this page Feb 2, 2018 · 32 revisions

Install script

On most systems you can install OpenFace using the ./install.sh script, if that does not work, use the following steps.

Advanced Ubuntu installation

For Unix based systems and different compilers, I included Cmake files for cross-platform and cross-IDE support.

This code has been tested on Ubuntu 14.04.1 with GCC, and on 15.10 with Clang 3.7.1.

You can also run the install.sh script for installing on Ubuntu 16.04 (it combines the following steps into one script)

Dependency installation

This requires cmake, OpenCV 3.1.0 (or newer), tbb and boost.

To acquire all of the dependencies follow the instructions pertaining to your Operating System:

Ubuntu 14.04

  1. Get newest GCC, done using:

    sudo apt-get update

    sudo apt-get install build-essential

  2. Cmake:

    sudo apt-get install cmake

  3. Get BLAS (for dlib)

    sudo apt-get install libopenblas-dev liblapack-dev

  4. OpenCV 3.1.0

    4.1 Install OpenCV dependencies:

     sudo apt-get install git libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-dev

 sudo apt-get install python-dev python-numpy libtbb2 libtbb-dev libjpeg-dev libpng-dev libtiff-dev libjasper-dev libdc1394-22-dev checkinstall

    4.2 Download OpenCV 3.1.0 from https://github.com/Itseez/opencv/archive/3.1.0.zip

     wget https://github.com/Itseez/opencv/archive/3.1.0.zip

    4.3 Unzip it and create a build folder:

     sudo unzip 3.1.0.zip
 cd opencv-3.1.0
 mkdir build
 cd build

    4.4 Build it using:

     cmake -D CMAKE_BUILD_TYPE=RELEASE -D CMAKE_INSTALL_PREFIX=/usr/local -D WITH_TBB=ON -D BUILD_SHARED_LIBS=OFF ..

    make -j2

    sudo make install

  5. Get Boost:

    sudo apt-get install libboost1.55-all-dev

    alternatively: sudo apt-get install libboost-all-dev

Ubuntu 15.10

  1. Get LLVM, Clang and libc++:

    1.1 Install LLVM and Clang dependencies:

    sudo apt-get update

    sudo apt-get install build-essential

    1.2 Install LLVM:

    sudo apt-get install llvm

    1.3 Install Clang along with relevant packages:

    • Check that the correct repository is enabled by inspecting '/etc/apt/sources.list':

      sudo gedit /etc/apt/sources.list

      If 'universe' is not included, modify the file so that it is:

      deb http://us.archive.ubuntu.com/ubuntu wily main universe

      After any such changes, update the system with:

      sudo apt-get update

    • Get Clang and libc++:

      sudo apt-get install clang-3.7 libc++-dev libc++abi-dev

  2. Get Cmake:

    sudo apt-get install cmake

  3. Get BLAS (for dlib)

    sudo apt-get install libopenblas-dev liblapack-dev

  4. OpenCV 3.1.0

    4.1 Install OpenCV dependencies:

     sudo apt-get install git libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-dev

 sudo apt-get install python-dev python-numpy libtbb2 libtbb-dev libjpeg-dev libpng-dev libtiff-dev libjasper-dev libdc1394-22-dev checkinstall

    4.2 Download OpenCV 3.1.0 from https://github.com/Itseez/opencv/archive/3.1.0.zip

     wget https://github.com/Itseez/opencv/archive/3.1.0.zip

    4.3 Unzip it and create a build folder:

     sudo unzip 3.1.0.zip
 cd opencv-3.1.0
 mkdir build
 cd build

    4.4 Build it using:

     cmake -D CMAKE_CXX_COMPILER=clang++ -D CMAKE_CXX_FLAGS="-std=c++11 -stdlib=libc++ -I/usr/include/libcxxabi" -D CMAKE_EXE_LINKER_FLAGS="-std=c++11 -stdlib=libc++ -lc++abi" -D CMAKE_BUILD_TYPE=RELEASE -D CMAKE_INSTALL_PREFIX=/usr/local -D OPENCV_EXTRA_MODULES_PATH=<path to opencv_contrib modules> -D BUILD_TIFF=ON -D WITH_V4L=ON -D WITH_GTK=ON -D BUILD_opencv_dnn=OFF -D WITH_TBB=ON ..

    make -j2

    sudo make install

  5. Get Boost 1.61.0:

    5.1 Install boost dependency:

    sudo apt-get install libbz2-dev

    5.2 Get the boost source:

    Download boost_1_61_0.tar.bz2.

    In the directory where you want to put the boost installation, execute:

    tar --bzip2 -xf /path/to/boost_1_61_0.tar.bz2

    5.3 Build and install:

     cd path/to/boost_1_61_0
 ./bootstrap.sh --with-toolset=clang --prefix=/usr/local
 ./b2 toolset=clang cxxflags="-std=c++11 -stdlib=libc++ -I/usr/include/libcxxabi" linkflags="-stdlib=libc++" --prefix=/usr/local -j 10 define=BOOST_SYSTEM_NO_DEPRECATED stage release
 sudo ./b2 install toolset=clang cxxflags="-std=c++11 -stdlib=libc++" linkflags="-stdlib=libc++" --prefix=/usr/local

Actual OpenFace installation

  1. Get OpenFace

    git clone https://github.com/TadasBaltrusaitis/OpenFace.git

  2. Create an out-of-source build directory to store the compiled artifacts:

     cd OpenFace
 mkdir build
 cd build

  3. Compile the code using instructions pertaining to your operating system:

    • Ubuntu 14.04

      cmake -D CMAKE_BUILD_TYPE=RELEASE ..

      make

    • Ubuntu 15.10

      cmake -D CMAKE_CXX_COMPILER=clang++ -D CMAKE_CXX_FLAGS="-std=c++11 -stdlib=libc++ -I/usr/include/libcxxabi" -D CMAKE_EXE_LINKER_FLAGS="-std=c++11 -stdlib=libc++ -lc++abi" -D CMAKE_BUILD_TYPE=RELEASE ..

      make

  4. Test it with

    • for videos:

      ./bin/FaceLandmarkVid -f "../samples/changeLighting.wmv" -f "../samples/2015-10-15-15-14.avi"

    • for images:

      ./bin/FaceLandmarkImg -fdir "../samples/" -wild

    • for multiple faces in videos:

      ./bin/FaceLandmarkVidMulti -f ../samples/multi_face.avi

    • for feature extraction (facial landmarks, head pose, AUs, gaze and HOG and similarity aligned faces):

      ./bin/FeatureExtraction -verbose -f "../samples/default.wmv"

Troubleshooting

X server

If you experience a problem with "cannot connect to X server" when trying to execute the tracker, a solution can be found here http://askubuntu.com/questions/64820/wkhtmltopdf-wkhtmltoimage-cannot-connect-to-x-server, to resolve run: apt-get install xvfb

Anaconda

When Anaconda is installed, somehow OpenCV finds the outdated GCC 4.x instead of GCC 5.4, according to: https://stackoverflow.com/questions/40322301/compile-opencv-3-on-ubuntu-16-04-linking-error-usr-lib-x86-64-linux-gnu-libsox

This results in OpenFace giving the error: usr/lib/x86_64-linux-gnu/libsoxr.so.0: undefined reference to `GOMP_parallel@GOMP_4.0' collect2: error: ld returned 1 exit status

Clone this wiki locally