rcabido/lsedataset

Spanish Sign Language dataset

lsedataset

Spanish Sign Language dataset

Getting Started

This project will consist of the construction of a video data set, subtitles of sign language people and their corresponding poses. The Openpose - The principal framework to detect the poses.

Prerequisites

To run this project, you need to install some version greater than Python 3 (with pip) and Git.

Ubuntu

Python

Open a Terminal
sudo apt update
sudo apt-get install python3.X 

  X is the number of version

Pip

Open a Terminal
sudo apt update
sudo apt-get install python3-pip

Git

Open a Terminal
sudo apt update
sudo apt-get install git

Windows

Python

1. Open a browser window and navigate to the Download page for Windows at python.org.
2. Underneath the heading at the top that says Python Releases for Windows, click on the link for the Latest Python 3 Release - Python 3.x.x. 
3. Scroll to the bottom and select either Windows x86-64 executable installer for 64-bit or Windows x86 executable installer for 32-bit. (See below.)
4. Run the installer

Pip

1. Before installing PIP, you’ll need to download get-pip.py: get-pip.py on pypa.io.
2. Open a cmd and navigate to download directory.
3. Run the command:
        python get-pip.py
4. Check the version:
        pip --version

Git

1. To install Git on Windows you will need to download the installer from the Git website:
2. Download the most current version for your operating system by double clicking on the package name:
3. Select Run to begin the installation:
4. Click Yes to continue:
5. Click Next to continue:

Installing

Clone or Download the project

1. Click to clone or download in the button that it is situated at top of this page.
2. Copy the link if you want clone it and go to 3. or download and unzip the file wherever you want.
3. Go to the directoy you want and execute:
      git clone **link** 
  Where the link is the result of second step

Install the Requeriments

1. Go to the directory 'lsedataset' 
2. Run the command:
      pip install -r Requirements.txt

Clone Openpose

Openpose is add to this project like a submodule, to clone it you have to run this commands:

1. Go to the directory 'lsedataset' 
  (if you want to go by commmand execute 'cd lsedataset' after clone this project)
2. Execute the following instrucctions:
      git submodule init
      git submodule update

Install Openpose GPU only mode

Install Cmake

  Windows
1. Install CMake GUI: Download and install the Latest Release of CMake Windows win64-x64 Installer from the CMake download website, called cmake-X.X.X-win64-x64.msi
2. Install Microsoft Visual Studio (VS) 2017 Enterprise with all C++-related flags when selecting the components to install.

  Ubuntu
1. Run the command:
      sudo apt-get install cmake-qt-gui.

OpenPose Configuration

1. Open CMake GUI and select the OpenPose directory as project source directory, and a non-existing or empty sub-directory called build where the Makefile files (Ubuntu) or Visual Studio solution (Windows) will be generated. If build does not exist, it will ask you whether to create it. Press Yes.

2. Press the Configure button, keep the generator in Unix Makefile (Ubuntu) or set it to your 64-bit Visual Studio version (Windows), and press Finish. Select the option only CPU in GPU model

3. If this step is successful, the Configuring done text will appear in the bottom box in the last line. Otherwise, some red text will appear in that same bottom box.

4. Press the Generate button and proceed to OpenPose Building. You can now close CMake.

For more information go to Openpose installation document:

• Openpose

Build Openpose

Ubuntu

Build the project by running the following commands.

cd build/ make -jnproc

Windows

Open the Visual Studio solution (Windows), called build/OpenPose.sln. Then, set the configuration from Debug to Release and press the green triangle icon to run it.

Installing Docker

If you want to build using Nvidia Gpu:

1. You have to install Cuda 10 or highest in the host system
2. Install the nvidia docker dependencies Nvidia.docker
3. To build the image run the command: docker build -t lsedataset_nvidia:latest -f gpu.Dockerfile .

Else if you want to build using only CPU

1. To build the image run the command: docker build -t lsedataset:latest -f cpu.Dockerfile .

Deployment

Tu run this project you have to go to the directory where contain the folder 'data' and 'redis' that you want to mount in the docker, then you have to execute one of the follow commands:

 docker run --rm --gpus all --name LSEdataSet --network mynet --mount type=bind,source="$(pwd)"/data,target=/openpose/VideosTFG  -it lsedataset_nvidia:latest
 docker run --rm --gpus all --name LSEdataSet --network mynet --mount type=bind,source="$(pwd)"/data,target=/openpose/VideosTFG  -it lsedataset:latest

There are some options in app that have integrated a date base redis, to run this options it necesary create a docker network and download redis image to run a container with some configuration:

docker network create mynet
docker pull redis
docker run -d --rm --name Redis -v /redis:/data --network mynet redis:latest

Choose the option that you desire

1. Will be generated a lsedataset formed by videos, subtitles and poses of all files that`s contains in the txt youtube urls file that you had introduced.
2. Will be generated the poses of all videos that contains the data's folder.
3. Will be generated a lsedataset formed by videos, subtitles and poses of all files that contains the first search Youtube page that it generated by a string wished.
4. Will be list all videos contais in data's folder.
5. Will be the path of a video-chapter-poses wished.
6. Will search in data base the model word-startVideoTime-endVideoTime-poses-video of a word wished.
7. Will list the words in data base.
8. Will add words that appear in a video contais in data's folder whised.
9. Delete a word whised in data base.
10. Exit

Authors

• Rubén Bernabé Menéndez

License

This project is licensed under the MIT License - see the LICENSE.md file for details