Visu

Editing tool built with automatic video analysis and visualisations of video features.

Installation Steps

The following installation steps were verified on Ubuntu 18.04. It is assumed that you have Python 3 installed, and it can be accessed via the python3 command.

1. npm install
2. Run the following commands to install FFmpeg 4
   1. sudo add-apt-repository ppa:jonathonf/ffmpeg-4
   2. sudo apt-get update
   3. sudo apt-get install ffmpeg
3. sudo apt-get install libavcodec-dev libavformat-dev libavdevice-dev
4. Run the following commands to install CUDA
   1. sudo add-apt-repository ppa:graphics-drivers/ppa
   2. sudo apt update
   3. sudo ubuntu-drivers autoinstall
   4. Reboot your machine
   5. sudo apt install nvidia-cuda-toolkit gcc-6
   6. Verify your CUDA install by running nvcc --version
5. Run sudo apt install python3-pip (if you don’t have pip3 installed already)
6. Run the following commands to install required packages
   1. cd src/python/scripts/mds/
   2. pip3 install -r requirements.txt
   3. cd ../yolo/
   4. pip3 install -r requirements.txt
   5. wget https://pjreddie.com/media/files/yolov3.weights
   6. mv yolov3.weights weights/
   7. cd ../hecate
7. Download OpenCV 3.4.7
8. Extract the above zip file and navigate to it in a new terminal
9. Run the following commands to install OpenCV
   1. mkdir build/
   2. cd build/
   3. cmake -D CMAKE_BUILD_TYPE=Release -D CMAKE_INSTALL_PREFIX=/usr/local ..
   4. make -j8
   5. sudo make install
   6. sudo ldconfig
10. Close this new terminal and go to your first terminal (should currently be inside the src/python_scripts/hecate/ folder)
11. Run the following commands to install hecate
    1. git clone https://github.com/yahoo/hecate.git
    2. cd hecate/
    3. make all
    4. make distribute
    5. mv distribute/ ..
    6. cd ..
    7. rm -rf hecate/
    8. Verify that hecate installed correctly by running ./distribute/bin/hecate
12. Run cd ../../../
13. Finally, run npm run electron:serve to start up Visu