Layout segmentation of ancient manuscripts Call for Competition

===================================

CALL FOR COMPETITION: SAM 2024  -   with prizes! $$$

Layout segmentation of ancient manuscripts

https://ai4ch.uniud.it/udiadscomp/

===================================

 

We are glad to announce SAM: International Competition on Few-Shot and
Many-Shot Layout Segmentation of Ancient Manuscripts, in conjunction
with the 18th International Conference on Document Analysis and
Recognition ICDAR 2024.

 

Competition Overview:

Layout segmentation is a critical aspect of Document Image Analysis,
particularly when it comes to ancient manuscripts. It consists in
decomposing the document in several regions representing title, main
text, paratext, etc.. We invite the research community to address this
task on U-DIADS-Bib, a novel dataset of fully-labelled ancient
manuscripts.

 

Competition Tasks:

We propose two separate tasks. Participants can try only one of them or both.

    Few-Shot Segmentation: participants are asked to develop a layout
    segmentation system using only three images for each manuscript as
    a training set

    Many-Shot Segmentation: participants have access to the full
    dataset (except for the private data that will be used for the
    final evaluation)

 

Important Dates:

    Beginning of Track 1: January 15, 2024
    Deadline of Track 1: March 3, 2024
    Beginning of Track 2: Match 4, 2024
    Deadline of Track 2: March 31, 2024

Prizes:

Winners of each task will be eligible for a cash prize of 300 EUR
sponsored by CVPL - Italian Association for Computer Vision,
Pattern Recognition and Machine Learning, IAPR Italian chapter.

 

For any additional information, please visit the website: 
https://ai4ch.uniud.it/udiadscomp/

 

Organizers:

Silvia Zottin, zottin.silvia@spes.uniud.it

Axel De Nardin, denardin.axel@spes.uniud.it

Claudio Piciarelli, claudio.piciarelli@uniud.it

Gian Luca Foresti, gianluca.foresti@uniud.it

Emanuela Colombi, emanuela.colombi@uniud.it

AI4CH - Artificial Intelligence for Cultural Heritage Lab, University of Udine.
https://ai4ch.uniud.it/