Computer Vision and Digital Manuscripts 

In an increasingly digital world, the automated analysis of ancient manuscripts is gaining importance for both academic research and cultural preservation. Document Analysis is becoming a key tool in addressing the unique challenges posed by historical manuscripts, which are often highly diverse in language, origin, writing style, layout, and condition.

To manage this complexity, advanced computational methods are needed to handle issues such as multilingual content, damaged texts, and varied codicological features. Beyond transcription, these technologies support digital preservation, scholarly editing, and integration into modern cultural platforms.

This workshop aims to connect experts in historical manuscripts with researchers in Computer Vision and AI, fostering interdisciplinary collaboration. By doing so, it seeks to develop innovative solutions that enhance the study, preservation, and accessibility of historical documents, ensuring their value is preserved for future generations.

Call of paper

The workshop encourages submissions related (but not limited) to the following topics:

• Handwritten Text Recognition (HTR) of historical documents

• Layout Analysis for historical documents

• Writer Identification in historical documents

• Historical Documents processing workflow

• Few-shot learning approaches for HTR and Layout Analysis

• Generative approaches to Historical Document Instance generation

• Computer Vision and Digital Codicology

• 3D reconstruction, visualization and modeling for Historical Documents

• Multispectral imaging for analyzing Historical Documents

• AI systems for Cultural Heritage Preservation

• Document Vision Question Answering

• Historical Document Classification

Important dates

• Paper Submission Deadline: 15 June 2025

• Authors Notification: 1 July 2025

• Camera ready papers due: 10 July 2025

Submission and review

We invite researchers to submit their original and unpublished work related to the workshop's theme. Authors can submit either full papers (12 including references ) or short papers (max 4 pages + reference), following the ICIAP 2025 formatting guidelines. Accepted full papers will published under the ICIAP 2025 workshop proceedings.

Submissions must be in PDF format and will be handled via the Microsoft CMT Conference system at the following address: TBA

All submissions should be compiled for double-blind review, adopt the standard main conference ICIAP 2025 template.

Each accepted paper must be covered by at least one registered author. Registration can be either for the full event at a regular rate or just for workshops and tutorials (2 days).