A Dataset for the Recognition of Historical and Handwritten Music Scores in Western Notation
Authors: Pau Torras, Jiří Mayer, Carles Badal, Martina Dvořáková, Markéta Herzanová Vlková, Gerard Asbert, Vojtěch Dvořák, Samuel Šomorjai, +2 more
Organizations: Computer Vision Center, Barcelona, Spain · Institute of Formal and Applied Linguistics, Charles University, Prague, Czechia · Department of Arts and Musicology, Universitat Autònoma de Barcelona · Moravian Library, Brno, Czechia · Department of Computer Science, Universitat Autònoma de Barcelona, Spain
A large amount of musical heritage has been digitised by memory institutions: libraries, museums, and archives. Nevertheless, the field of Optical Music Recognition (OMR) has struggled with making this music machine-readable, despite advances in deep learning, mostly because no datasets for training systems in realistic conditions were available. The MusiCorpus dataset aims to remedy this situation by providing 1,309 pages of historical sheet music, primarily handwritten, with MusicXML transcriptions and symbol annotations. It is the largest dataset of handwritten music to date and the first dataset containing a realistic and representative sample of musical document collections from memory institutions, suitable for training and evaluating both end-to-end and object detection-based OMR systems and comparing their performance.