Tutorial

This documentation is aimed at Digital Humanities specialists who want to transcribe handwritten or printed documents using our Arkindex platform.

In this section you will discover how to run a full project using Arkindex and Callico.

We will cover the following topics:

  1. Import openly available datasets into Arkindex,

  2. Create ground truth annotations for segmentation using Callico,

  3. Train a new segmentation model with your own annotated data,

  4. Create ground truth annotations for transcription using Callico,

  5. Train a new transcription model with your own annotated data,

  6. Run your own models on any image you wish to segment and transcribe,

  7. Export the segmentation and transcription results out of Arkindex, and in a PAGE XML format,

  8. Conclusion and limitations.

The tutorial overview that will introduce this section’s content is a good place to start.