Models from paper "MuSViT: A Foundation Vision Model for Sheet Music Representation"
AI & ML interests
Computer Vision, Optical Music Recognition, Audio-to-Score transcription
Recent Activity
Papers
MuSViT: A Foundation Vision Model for Sheet Music Representation
Optical Music Recognition of Jazz Lead Sheets
All the available datasets for the Audio-to-Score (A2S) tasks
The GRANDSTAFF collection contains 53,882 single-system piano scores in common western modern notation. The audio recordings were synthesized.
The Quartets collection comprises three datasets of string quartets by Haydn, Mozart, and Beethoven, retrieved from the Humdrum repository
Datasets used in OPTICAL MUSIC RECOGNITION OF JAZZ LEAD SHEETS - ISMIR 2025 - Daejeon, Korea
Models from paper "MuSViT: A Foundation Vision Model for Sheet Music Representation"
All the available datasets for multimodal Optical Music Recognition (OMR) and Audio-to-Score (A2S)
All the available datasets for the Audio-to-Score (A2S) tasks
Pianoform datasets
The GRANDSTAFF collection contains 53,882 single-system piano scores in common western modern notation. The audio recordings were synthesized.
Weights of the Sheet Music Transformer project
The Quartets collection comprises three datasets of string quartets by Haydn, Mozart, and Beethoven, retrieved from the Humdrum repository
All the available datasets for OMR.
Datasets used in OPTICAL MUSIC RECOGNITION OF JAZZ LEAD SHEETS - ISMIR 2025 - Daejeon, Korea
Datasets for Aligned Music Notation and Lyrics Transcription