Juan P. Martinez-Esteso
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Juan P. Martinez-Esteso has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Antonio Javier Gallego
University Institute for Computing Research, University of Alicante, Alicante, Spain
6 shared publications - Francisco J. Castellanos
University of Alicante
6 shared publications - Alejandro Galán-Cuenca
University of Alicante
3 shared publications - Adrian Rosello
University of Alicante
2 shared publications Jorge Calvo-Zaragoza
1 shared publication- Jorge Calvo-Zaragoza
University of Alicante
1 shared publication
Latest Publications
- 9 May 2026
- Published by Springer Nature
- p. 325-330
Can Patch Selection Heuristics Enhance Layout Analysis of Music Scores?
- 9 May 2026
- Published by Springer Nature
- p. 325-330
Can Patch Selection Heuristics Enhance Layout Analysis of Music Scores?
- Open Access
- 16 December 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 305
Enhancing maritime search and rescue: Incremental unsupervised domain adaptation with synthetic data and pseudo-labeling
- 16 December 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 305
Open AccessEnhancing maritime search and rescue: Incremental unsupervised domain adaptation with synthetic data and pseudo-labeling
- Open Access
- 20 June 2025
- Published by Springer Nature in International Journal on Document Analysis and Recognition (IJDAR)
- p. 1-16
Enhancing music score analysis with Monte Carlo dropout: a probabilistic approach to staff-region detection
Layout Analysis (LA) is a critical process for detecting and isolating different components within a scanned document, allowing for more straightforward and precise processing of each part independently. In Optical Music Recognition (OMR), LA is essential for identifying and extracting music staves, which enables effective music notation recognition and processing. While the literature includes several studies exploring methods for staff retrieval, there remains room for improvement in terms of robustness and accuracy. In this work, we introduce a methodology that integrates Monte Carlo Dropout (MCD) into a neural network model in order to improve reliability in staff retrieval from scanned sheet music. Our approach leverages multiple non-deterministic predictions using standard dropout layers during inference and aggregates them through pixel-level combination policies. We extend the MCD technique, originally designed for classification and regression tasks using averaged predictions, to the LA task and introduce new combination strategies: maximum and voting criteria. Experiments on three diverse music score corpora, including printed and handwritten documents, demonstrated the effectiveness of our approach. The averaging and voting (with 25% and 50% of votes) criteria reduced the relative error by 63.6% compared to the baseline and achieved a 32.1% improvement over state-of-the-art methods. Our methodology notably enhanced detection accuracy without requiring modifications to the neural architecture, especially at the edges of staves, where conventional models tend to show higher error rates.
- 20 June 2025
- Published by Springer Nature in International Journal on Document Analysis and Recognition (IJDAR)
- p. 1-16
Open AccessEnhancing music score analysis with Monte Carlo dropout: a probabilistic approach to staff-region detection
Layout Analysis (LA) is a critical process for detecting and isolating different components within a scanned document, allowing for more straightforward and precise processing of each part independently. In Optical Music Recognition (OMR), LA is essential for identifying and extracting music staves, which enables effective music notation recognition and processing. While the literature includes several studies exploring methods for staff retrieval, there remains room for improvement in terms of robustness and accuracy. In this work, we introduce a methodology that integrates Monte Carlo Dropout (MCD) into a neural network model in order to improve reliability in staff retrieval from scanned sheet music. Our approach leverages multiple non-deterministic predictions using standard dropout layers during inference and aggregates them through pixel-level combination policies. We extend the MCD technique, originally designed for classification and regression tasks using averaged predictions, to the LA task and introduce new combination strategies: maximum and voting criteria. Experiments on three diverse music score corpora, including printed and handwritten documents, demonstrated the effectiveness of our approach. The averaging and voting (with 25% and 50% of votes) criteria reduced the relative error by 63.6% compared to the baseline and achieved a 32.1% improvement over state-of-the-art methods. Our methodology notably enhanced detection accuracy without requiring modifications to the neural architecture, especially at the edges of staves, where conventional models tend to show higher error rates.
- Open Access
- 7 November 2024
- Published by Elsevier in Engineering Applications of Artificial Intelligence
- Vol. 139
On the use of synthetic data for body detection in maritime search and rescue operations
- 7 November 2024
- Published by Elsevier in Engineering Applications of Artificial Intelligence
- Vol. 139
Open AccessOn the use of synthetic data for body detection in maritime search and rescue operations
- 11 September 2024
- Published by Springer Nature
- p. 58-75
A Region-Based Approach for Layout Analysis of Music Score Images in Scarce Data Scenarios
- 11 September 2024
- Published by Springer Nature
- p. 58-75
A Region-Based Approach for Layout Analysis of Music Score Images in Scarce Data Scenarios
- 25 June 2023
- Published by Springer Nature
- p. 158-169
Test-Time Augmentation for Document Image Binarization
- 25 June 2023
- Published by Springer Nature
- p. 158-169
Test-Time Augmentation for Document Image Binarization