Optical Measure Recognition in Common Music Notation. Vigliensoni, G., Gregory, B., & Fujinaga, I. In Souza Britto Jr., Alceu de, Gouyon, F., & Dixon, S., editors, Proceedings of the 14th International Society for Music Information Retrieval Conference, ISMIR 2013, Curitiba, Brazil, November 4–8, 2013, pages 125–130, 2013.
Paper abstract bibtex 13 downloads This paper presents work on the automatic recognition of measures in common Western music notation scores using optical music recognition techniques. It is important to extract the bounding boxes of measures within a music score to facilitate some methods of multimodal navigation of music catalogues. We present an image processing algorithm that extracts the position of barlines on an input music score in order to deduce the number and position of measures on the page. An open-source implementation of this algorithm is made publicly available. In addition, we have created a ground-truth dataset of 100 images of music scores with manually annotated measures. We conducted several experiments using different combinations of values for two critical parameters to evaluate our measure recognition algorithm. Our algorithm obtained an f-score of 91 percent with the optimal set of parameters. Although our implementation obtained results similar to previous approaches, the scope and size of the evaluation dataset is significantly larger.
@inproceedings{Vigliensoni_2013,
abstract = {This paper presents work on the automatic recognition of measures in common Western music notation scores using optical music recognition techniques. It is important to extract the bounding boxes of measures within a music score to facilitate some methods of multimodal navigation of music catalogues. We present an image processing algorithm that extracts the position of barlines on an input music score in order to deduce the number and position of measures on the page. An open-source implementation of this algorithm is made publicly available. In addition, we have created a ground-truth dataset of 100 images of music scores with manually annotated measures. We conducted several experiments using different combinations of values for two critical parameters to evaluate our measure recognition algorithm. Our algorithm obtained an f-score of 91 percent with the optimal set of parameters. Although our implementation obtained results similar to previous approaches, the scope and size of the evaluation dataset is significantly larger.},
author = {Vigliensoni, Gabriel and Gregory, Burlet and Fujinaga, Ichiro},
title = {Optical Measure Recognition in Common Music Notation},
url = {http://ismir2013.ismir.net/wp-content/uploads/2013/09/207_Paper.pdf},
pages = {125–130},
editor = {{Souza Britto Jr., Alceu de} and Gouyon, Fabien and Dixon, Simon},
booktitle = {Proceedings of the 14th International Society for Music Information Retrieval Conference, ISMIR 2013, Curitiba, Brazil, November 4–8, 2013},
year = {2013}
}
Downloads: 13
{"_id":"rdjhZXDdXL6bi6s9E","bibbaseid":"vigliensoni-gregory-fujinaga-opticalmeasurerecognitionincommonmusicnotation-2013","author_short":["Vigliensoni, G.","Gregory, B.","Fujinaga, I."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","abstract":"This paper presents work on the automatic recognition of measures in common Western music notation scores using optical music recognition techniques. It is important to extract the bounding boxes of measures within a music score to facilitate some methods of multimodal navigation of music catalogues. We present an image processing algorithm that extracts the position of barlines on an input music score in order to deduce the number and position of measures on the page. An open-source implementation of this algorithm is made publicly available. In addition, we have created a ground-truth dataset of 100 images of music scores with manually annotated measures. We conducted several experiments using different combinations of values for two critical parameters to evaluate our measure recognition algorithm. Our algorithm obtained an f-score of 91 percent with the optimal set of parameters. Although our implementation obtained results similar to previous approaches, the scope and size of the evaluation dataset is significantly larger.","author":[{"propositions":[],"lastnames":["Vigliensoni"],"firstnames":["Gabriel"],"suffixes":[]},{"propositions":[],"lastnames":["Gregory"],"firstnames":["Burlet"],"suffixes":[]},{"propositions":[],"lastnames":["Fujinaga"],"firstnames":["Ichiro"],"suffixes":[]}],"title":"Optical Measure Recognition in Common Music Notation","url":"http://ismir2013.ismir.net/wp-content/uploads/2013/09/207_Paper.pdf","pages":"125–130","editor":[{"firstnames":[],"propositions":[],"lastnames":["Souza Britto Jr., Alceu de"],"suffixes":[]},{"propositions":[],"lastnames":["Gouyon"],"firstnames":["Fabien"],"suffixes":[]},{"propositions":[],"lastnames":["Dixon"],"firstnames":["Simon"],"suffixes":[]}],"booktitle":"Proceedings of the 14th International Society for Music Information Retrieval Conference, ISMIR 2013, Curitiba, Brazil, November 4–8, 2013","year":"2013","bibtex":"@inproceedings{Vigliensoni_2013,\n abstract = {This paper presents work on the automatic recognition of measures in common Western music notation scores using optical music recognition techniques. It is important to extract the bounding boxes of measures within a music score to facilitate some methods of multimodal navigation of music catalogues. We present an image processing algorithm that extracts the position of barlines on an input music score in order to deduce the number and position of measures on the page. An open-source implementation of this algorithm is made publicly available. In addition, we have created a ground-truth dataset of 100 images of music scores with manually annotated measures. We conducted several experiments using different combinations of values for two critical parameters to evaluate our measure recognition algorithm. Our algorithm obtained an f-score of 91 percent with the optimal set of parameters. Although our implementation obtained results similar to previous approaches, the scope and size of the evaluation dataset is significantly larger.},\n author = {Vigliensoni, Gabriel and Gregory, Burlet and Fujinaga, Ichiro},\n title = {Optical Measure Recognition in Common Music Notation},\n url = {http://ismir2013.ismir.net/wp-content/uploads/2013/09/207_Paper.pdf},\n pages = {125–130},\n editor = {{Souza Britto Jr., Alceu de} and Gouyon, Fabien and Dixon, Simon},\n booktitle = {Proceedings of the 14th International Society for Music Information Retrieval Conference, ISMIR 2013, Curitiba, Brazil, November 4–8, 2013},\n year = {2013}\n}\n\n","author_short":["Vigliensoni, G.","Gregory, B.","Fujinaga, I."],"editor_short":["Souza Britto Jr., Alceu de","Gouyon, F.","Dixon, S."],"key":"Vigliensoni_2013","id":"Vigliensoni_2013","bibbaseid":"vigliensoni-gregory-fujinaga-opticalmeasurerecognitionincommonmusicnotation-2013","role":"author","urls":{"Paper":"http://ismir2013.ismir.net/wp-content/uploads/2013/09/207_Paper.pdf"},"metadata":{"authorlinks":{}},"downloads":13,"html":""},"bibtype":"inproceedings","biburl":"https://raw.githubusercontent.com/music-encoding/music-encoding.github.io/master/resources/mei_bibliography.bib","dataSources":["nAQEJTcugBFY2RYhG","sKksfkgR6E2tYsJGP"],"keywords":[],"search_terms":["optical","measure","recognition","common","music","notation","vigliensoni","gregory","fujinaga"],"title":"Optical Measure Recognition in Common Music Notation","year":2013,"downloads":13}