Gamera Versus Aruspix: Two Optical Music Recognition Approaches

Optical music recognition (OMR) applications are predominantly designed for common music notation and as such, are inherently incapable of adapting to specialized notation forms within early music. Two OMR systems, namely Gamut (a Gamera application) and Aruspix, have been proposed for early music. In this paper, we present a novel comparison of the two systems, which use markedly different approaches to solve the same problem, and pay close attention to the performance and learning rates of both applications. In order to obtain a complete comparison of Gamut and Aruspix, we evaluated the core recognition systems and the pitch determination processes separately. With our experiments, we were able to highlight the advantages of both approaches as well as causes of problems and possibilities for future improvements.

[1]  Christoph Wolff.  Repertoire International Des Sources Musicales (RISM) , 2007 .

[2]  Ichiro Fujinaga,et al.  A Comparative Survey of Image Binarisation Algorithms for Optical Recognition on Degraded Musical Sources , 2007, ISMIR.

[3]  Michael Droettboom.  Correcting broken characters in the recognition of historical printed documents , 2003, 2003 Joint Conference on Digital Libraries, 2003. Proceedings..

[4]  Ichiro Fujinaga,et al.  A Comparative Study of Staff Removal Algorithms , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[5]  Klaus Keil.  Repertoire International Des Sources Musicales (RISM) , 2009 .

[6]  Ichiro Fujinaga,et al.  USING HIDDEN MARKOV MODELS , 2007 .

[7]  Laurent Pugin,et al.  Optical Music Recognition of Early Typographic Prints using Hidden Markov Models , 2006 .

[8]  Susan Ella George,et al.  Visual Perception of Music Notation: On-Line and Off-Line Recognition , 2004 .

[9]  Ichiro Fujinaga,et al.  Goal-directed evaluation for the improvement of optical music recognition on early music prints , 2007, JCDL '07.

[10]  Ichiro Fujinaga,et al.  Using the Gamera framework for the recognition of cultural heritage materials , 2002, JCDL '02.

[11]  Timothy C. Bell,et al.  The Challenge of Optical Music Recognition , 2001, Comput. Humanit..

[12]  Ichiro Fujinaga,et al.  Gamera: Optical music recognition in a new shell , 2002, ICMC.

[13]  João Rogério Caldas Pinto,et al.  Ancient Music Recovery for Digital Libraries , 2000, ECDL.

[14]  Pierfrancesco Bellini,et al.  Assessing Optical Music Recognition Tools , 2007, Computer Music Journal.

[15]  Donald Byrd,et al.  Prospects for Improving OMR with Multiple Recognizers , 2006, ISMIR.

[16]  Christophe Dalitz,et al.  Using the Gamera Framework for Building a Lute Tablature Recognition System , 2005, ISMIR.

[17]  Ichiro Fujinaga,et al.  Staff Detection and Removal , 2004 .

[18]  Laurent Pugin,et al.  Optical Music Recognitoin of Early Typographic Prints using Hidden Markov Models , 2006, ISMIR.