eRS is an emotion recognition system dedicated to researchers, whose purpose is to help them with emotion recognition for movies tasks.
Python JavaScript CoffeeScript HTML CSS Shell Batchfile
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
ers_backend
ers_frontend
scripts
LICENSE
README.md
install_ers_ubuntu.sh
requirements.txt
test_server.png

README.md

Project homepage: http://ers.humantech.institute/#/

eRS

eRS is an emotion recognition system dedicated to researchers, whose purpose is to help them with emotion recognition in movies tasks.

To see what you can do with eRS, have a look to http://ers.humantech.institute/#/presentation

Installation

eRS works on Linux and Mac OS X. (it is planned to dockerize it soon)

The script install_ers_ubuntu.sh is provided to install all dependencies on Ubuntu (tested on Ubuntu 14.10). Run it with sudo.

If you want to manually install the system, follow these instructions:

Dependencies

Install main dependencies (many of these dependencies are available on Mac OS X with macport or brew):

Then, install python packages:

sudo pip install -r requirements.txt

Then, install web packages:

cd ers_frontend
sudo npm install
bower install

Database

Create a database with information specified in ers_backend/ers_backend/settings.py in the Database configuration part. Default are:

  • Database name : ers_backend_db
  • User : root
  • Password : root

Control that the user root has all privileges on this database.

Running eRS

You can use the provided scripts to run the required services. Be sure to call each script from this path, and not from the script folder.

1. Celery workers (using Redis broker)

./scripts/run_redis.sh
./scripts/rund_celery.sh

2. Django REST server

./scripts/run_django.sh

3. SwampDragon WebSockets server

./scripts/run_swampdragon.sh

4. Frontend web server

./scripts/run_brunch.sh

Test eRS

Backend

We don't provide test videos for legal issues. You have to add test videos manually (use small .avi videos like 30 seconds maximum).

First, create the following folders:

$ers_installation_path$/datasets/test_dataset
$ers_installation_path$/datasets/test_dataset/video
$ers_installation_path$/datasets/test_dataset/video/anger
$ers_installation_path$/datasets/test_dataset/video/neutral

Then, add test videos:

$ers_installation_path$/datasets/test_dataset/video/1_shot.avi
$ers_installation_path$/datasets/test_dataset/video/anger/anger.avi
$ers_installation_path$/datasets/test_dataset/video/neutral/neutral.avi

1_shot.avi should contain 1 and only 1 shot boundary (an easy one, hard cut).

Then, launch the Django project unit tests to verify that the backend is working:

cd ers_backend
python manage.py test

Frontend

Open your web browser and go to http://localhost:3333/#/home to verify that the web application is accessible.

Then go to the page "Test server", and click on each buttons to test the connection with the different backend services. Alt text

License

This software is released under the MIT License.