Low-Data Classification of Historical Music Manuscripts: A Few-Shot Learning Approach

TL;DR

This paper presents a self-supervised learning framework for classifying musical symbols in historical manuscripts, enabling effective recognition with minimal labeled data and aiding digital preservation of musical heritage.

Contribution

It introduces a novel self-supervised CNN approach optimized for historical music manuscripts, demonstrating high accuracy with limited labeled samples.

Findings

Achieved 87.66% classification accuracy.

Effective use of self-supervised learning on unlabelled data.

Validated multiple classification methods including SVM and prototypical networks.

Abstract

In this paper, we explore the intersection of technology and cultural preservation by developing a self-supervised learning framework for the classification of musical symbols in historical manuscripts. Optical Music Recognition (OMR) plays a vital role in digitising and preserving musical heritage, but historical documents often lack the labelled data required by traditional methods. We overcome this challenge by training a neural-based feature extractor on unlabelled data, enabling effective classification with minimal samples. Key contributions include optimising crop preprocessing for a self-supervised Convolutional Neural Network and evaluating classification methods, including SVM, multilayer perceptrons, and prototypical networks. Our experiments yield an accuracy of 87.66\%, showcasing the potential of AI-driven methods to ensure the survival of historical music for future generations through advanced digital archiving techniques.