Layout Parser Annotation Service

Usage

We package all the layout annotation service (the annotation interface and active learning modeling server) inside docker containers. The installation process is very straightforward and simple:

1. Install Docker on your computer, following the official instructions.
2. Clone this repository to your computer.

   git clone git@github.com:Layout-Parser/annotation-service.git
   cd annotation-service

3. Configure the annotation folders (see details in the section below) and start the docker container

   DATA=./data CONFIG=labeling-config.xml MODEL=model.py docker-compose up --build -d

4. Go to localhost:8080 and start annotating.
5. Export the completed annotations via Label-Studio's export function, or you can find the annotation folder directly at labeled.

Configuration

In the 3rd command, the environmental variables DATA, CONFIG, and MODEL are used to set the labeling data directory, Label Studio configuration file, and ML backend model file, respectively.

• DATA is for the folder containing all the images for labeling. By default, DATA=./data.
• CONFIG is the configuration file for initializing the label-studio interface. The default value is CONFIG=horizontal-layout.xml, and you could find more examples in labeling/configs.
• MODEL is for the script that generates the model prediction. The default value is MODEL=model.py.

TODO

• [ ] Enable the Active Learning Detectron2 model backend.