Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Skip to content
/ eva Public
forked from antingshen/BeaverDam

EVA is a web-based tool for efficient annotation of videos and image sequences and has an additional tracking capabilities

License

Notifications You must be signed in to change notification settings

Ericsson/eva

 
 

Repository files navigation

EVA: Video Annotation Tool

EVA is a web-based tool for efficient annotation of videos and image sequences. It is a re-design of BeaverDam with additional tracking capabilities. The annotation is done on a bounding box level and the labels can be exported in YOLO or Pascal VOC format.

Setup

Windows:

Requirements:

Install:

git clone https://github.com/Ericsson/eva.git
cd eva
python -m pip install --upgrade pip
pip install virtualenv
virtualenv venv
venv\Scripts\activate.bat
pip install -r requirements.txt
python manage.py preparetracker
python manage.py migrate
python manage.py collectstatic

Extract ffmpeg archive and copy ffmpeg.exe, from the bin folder, to the root of the tool\ folder e.g. if you clone the repository as eva place the ffmpeg.exe inside eva\.

Start the app by running the start.bat file. In Chrome or Firefox go to http://127.0.0.1:8000/.

Linux:

Requirements:

  • Python 3.6, redis, ffmpeg, g++.
  • Browser: Google-Chrome Version > 49.0, Firefox Version > 45.0

Install:

git clone https://github.com/Ericsson/eva.git
cd eva
pip3 install virtualenv
python3 -m virtualenv venv
. venv/bin/activate
pip install --upgrade pip
pip install -r requirements.txt
python manage.py preparetracker
python manage.py migrate
python manage.py collectstatic

Start the app by running the start.sh file. In Chrome or Firefox go to http://127.0.0.1:8000/.

Docker:

Run using docker-compose

First run the following commands to initialize the tool. These only have to be run once, but if the tool is updated they should be repeated.

docker-compose build
docker-compose run eva python3 manage.py migrate
docker-compose run eva python3 manage.py collectstatic

Start the app by running:

docker-compose up

In Chrome or Firefox open the http link: http://127.0.0.1:8000/

Changing parameters:

You can change the parameters of the tracker by editing cfg/KCF_config.yml.

Acknowledgement:

Thanks to Ludwig Thaung for his contribution in building the EVA tool.

About

EVA is a web-based tool for efficient annotation of videos and image sequences and has an additional tracking capabilities

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 44.0%
  • JavaScript 42.3%
  • HTML 9.0%
  • CSS 3.6%
  • Shell 0.8%
  • Dockerfile 0.2%
  • Batchfile 0.1%