- Understanding and analyzing features of large music score corpora is a complex problem and provides numerous research opportunities within a growing community
- Detecting similarities and differences of a large score set is an interesting task for domain experts
- Manually investigating and extracting information and features from music scores is tedious and time-consuming
- The objective is to develop an interactive visual music score corpora explorer which supports users in getting an overview a large music score dataset available in MusicXML
- Implementation of user-steered analysis methods based on new visualizations to identify musical patterns of and between different low- and high-level features
- How can we visualize relevant information of large music score collections?
- How can musical features be visually encoded effectively?
- What features are interesting for music analysts and how to support effective user interaction?
- Automatically extracting semantic information that is relevant to analysts is difficult or even impossible.
- Perform a literature research to identify gaps of state-of-the-art visualization solution that support music score collection analysis (recommended: within the accompanying seminar)
- Examine and identify existing approaches and analysis models
- Work with a large dataset of music scores based on the established MusicXML standard format using the Python Library music21
- Implement a visualization pipeline that processes and transforms raw MusicXML files into interactive visualization models supporting user-steered analysis
- Help analysts to focus on features of interest ranging from meta-features (e.g., composer, epoch, genre) to low-level features (e.g., dynamics, articulation, instructions, tonality)
- Deep Interest in learning about music, its features, and state-of-the-art formats
- Basic knowledge about information visualization (e.g., Visual Variables, Gestalt Laws, Visual Information-Seeking Mantra, Linking & Brushing, etc.)
- Readiness to do creative research work independently
- Basic knowledge of Git (Issues, Feature Branches, Documentation, etc.)
- Scope: Bachelor / Master
- Project/Thesis Duration: 3 months/3 months (Bachelor), 6 months / 6 months (Master)
- Start: immediately
- M. Good. MusicXML for notation and analysis. The virtual score: representation, retrieval, restoration 12:113-124, 2001.
- W. Chan, and Qu Huamin. A Report on Musical Structure Visualization. Leonardo, Department of Computer Science and Engineering Hong Kong University of Science and Technology Clear Water Bay, Kowloon, Hong Kong, 2007.
- T. Bergstrom, K. Karahalios, and J. C. Hart. Isochords: visualizing structure in music. In Proceedings of Graphics Interface 2007. ACM, 2007.
- Khulusi, R., Kusnick, J., Focht, J., & Jänicke, S. (2020). musiXplora: Visual Analysis of a Musicological Encyclopedia. In VISIGRAPP (3: IVAPP) (pp. 76-87).
- Miller, M., Schäfer, H., Kraus, M., Leman, M., Keim, D., & El-Assady, M. (2019). Framing Visual Musicology through Methodology Transfer. arXiv preprint arXiv:1908.10411.
- Khulusi, R., Kusnick, J., Meinecke, C., Gillmann, C., Focht, J., & Jänicke, S. (2020, January). A survey on visualizations for musical data. In Computer Graphics Forum.
- Corrêa, D. C., & Rodrigues, F. A. (2016). A survey on symbolic data-based music genre classification. Expert Systems with Applications, 60, 190-210.
- Holder, E., Tilevich, E., & Gillick, A. (2015, October). Musiplectics: computational assessment of the complexity of music scores. In 2015 ACM International Symposium on New Ideas, New Paradigms, and Reflections on Programming and Software (Onward!) (pp. 107-120).
- DeCurtins, M. (2020). Clef: An Extensible, Experimental Framework for Music Information Retrieval (Doctoral dissertation).
- Jones, J., de Siqueira Braga, D., Tertuliano, K., & Kauppinen, T. (2017, August). MusicOWL: the music score ontology. In Proceedings of the International Conference on Web Intelligence (pp. 1222-1229).