Skip to content

CrowdCurio/image-annotator

Repository files navigation

CrowdCurio Image Annotator Library

The CrowdCurio Image Annotation Library implements counting (classification) and transcription tasks for images.

An screenshot of the Image Annotator.

"The interface is so mature." - Hemant Surale.

Features

  • Support for counting and (letter-by-letter) transcription tasks.
  • Support for alternating between practice tasks and required tasks.
  • Zoom functionality.
  • Fullscreen functionality.
  • Integrated support for CrowdCurio.

Build Process

We use Browserify and Uglify in our build processes. Both tools can be installed with NPM.

npm install -g browserify

npm install -g uglify-js

To build the script bundle without minification, run:

browserify lib/main.js -o bundle.js

To build with minification, run:

browserify lib/main.js | uglifyjs bundle.js

Usage

The UI is controlled by a set of parameters upon instantiation. Here's an example of the configuration and its usage:

// transcription configuration
var config = {
    mode: 'transcription',  # required
    language: 'greek',      # required
    practice: {
        active: false
    }
}

// counting configuration
window.config = {
            mode: 'counting',
            labels: [{
                name: 'Tau',
                examples: [
                    'url1',
                    'url2',
                    'url3'
                ]
            }, {
                name: 'Nu',
                examples: [
                    'url1',
                    'url2',
                    'url3'
                ]
            }],
            practice: {
                active: true,
                feedback: true
            }
        };

// define a new annotator
var annotator = new ImageAnnotator();
annotator.initialize(config);

Contact

Alex Williams, University of Waterloo