Experimenting active and sequential learning in a medieval music manuscript

TL;DR

This study explores active and sequential learning techniques for optical music recognition in medieval manuscripts, demonstrating that uncertainty-based AL may not always be effective in data-scarce scenarios.

Contribution

It introduces a preliminary application of AL and SL methods to medieval music manuscripts using YOLOv8, highlighting limitations of uncertainty-based AL in such contexts.

Findings

Uncertainty-based AL was ineffective on the medieval manuscript dataset.

Comparable accuracy to full supervision achieved with fewer labeled samples.

Highlights need for more usable AL methods in data-scarcity scenarios.

Abstract

Optical Music Recognition (OMR) is a cornerstone of music digitization initiatives in cultural heritage, yet it remains limited by the scarcity of annotated data and the complexity of historical manuscripts. In this paper, we present a preliminary study of Active Learning (AL) and Sequential Learning (SL) tailored for object detection and layout recognition in an old medieval music manuscript. Leveraging YOLOv8, our system selects samples with the highest uncertainty (lowest prediction confidence) for iterative labeling and retraining. Our approach starts with a single annotated image and successfully boosts performance while minimizing manual labeling. Experimental results indicate that comparable accuracy to fully supervised training can be achieved with significantly fewer labeled examples. We test the methodology as a preliminary investigation on a novel dataset offered to the community by the Anonymous project, which studies laude, a poetical-musical genre spread across Italy during the 12th-16th Century. We show that in the manuscript at-hand, uncertainty-based AL is not effective and advocates for more usable methods in data-scarcity scenarios.