Region-based layout analysis of music score images

作者：

Highlights：

• First study of object-detection models for Layout Analysis (LA) on music scores.

• Analysis of the suitability of the well-known mAP metric in this context.

• Proposal of a new data-generation method for building robust LA models.

• Goal-directed experiments of the influence of LA on the eventual transcription.

摘要

•First study of object-detection models for Layout Analysis (LA) on music scores.•Analysis of the suitability of the well-known mAP metric in this context.•Proposal of a new data-generation method for building robust LA models.•Goal-directed experiments of the influence of LA on the eventual transcription.

论文关键词：Optical Music Recognition,Layout Analysis,Image augmentation,Object detection

论文评审过程：Received 25 January 2022, Revised 28 June 2022, Accepted 16 July 2022, Available online 21 July 2022, Version of Record 30 July 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118211