Deep Learning for a Self-adaptive Optical Music Recognition System

Publié le ven 18/03/2022 - 18:52
Type de contrat
Fixed-term contract: up to 18 months
Corps / Catégorie
Research Engineer or Postdoc
Equipe de recherche
Contexte
IRISA - Intuidoc

IRISA is a joint research center for Informatics, including Robotics and Image and Signal Processing. 850 people, 40 teams, explore the world of digital sciences to find applications in healthcare, ecology-environment, cyber-security, transportation, multimedia, and industry... INSA Rennes is one of the 8 trustees of IRISA.

The Intuidoc team (https://www.irisa.fr/intuidoc) conducts researches on the topic of document image recognition. Since many years, the team proposes a system, called DMOS-PI method, for document structure analysis of documents. This DMOS-PI method is used for document recognition, or field extraction in archive documents, handwritten contents damaged documents (musical scores, archives, newspapers, letters, electronic schema, …).

Collabscore project

Collabscore is a project founded by ANR (French Research National Agency), led by the CNAM. The goal is to study ancient scores provided by the BNF (Bibliothèque National de France) and Royaumont foundation. Collabscore is a multidisciplinary project. The first task consists in improving OMR (Optical Music Recognition) results using learning techniques. The second action will focus on methods for automatic alignment of the scored score with other multimodal sources. The last one will set up demonstrators based on notated scores at two of the project partners, representative, in various ways, of institutions in charge of musical heritage collections (BnF and Fondation Royaumont). Intuidoc team focuses on the first task of musical score recognition.
Mission

The engineer/post-doctoral fellow will work on the conception of an OMR system. Based on previous works of our research team, the goal of this position is to enrich an existing system (DMOS-PI) to get a complete self-adaptative OMR system for historical orchestra scores. The tasks are mainly:

  • Generate unsupervised data for training musical symbols recognizers using Generative Adversarial Networks (GAN). For this task, one of our previous works, called Isolating-GAN, can be used to detect music symbols in an unsupervised way;
  • Design a musical symbols recognizer based on neural networks;
  • Build a self-adaptive system which can be robust on new partitions, where only few annotated data are available in training.

Machine Learning methods, especially Deep learning-based approaches (GAN, RCNN, SSD...), will be used to solve some of the tasks.

Profil / Compétences
- PhD in computer science or Master degree
- Experience in document recognition or statistical analysis.
- Knowledge in deep learning with an experience with at least one library dedicated to deep learning (Keras, Tensorflow, Pytorch) are expected.
Diplôme requis
Master degree or PhD in computer science.
Lieu de travail
IRISA Rennes
Date prévisionnelle d'embauche
Date limite de candidature
Durée du contrat (en mois)
18
Quotité
100%
Salaire brut mensuel
Up to €36 000 gross annual salary (according to experience), with social security benefits
Candidater
Candidates should contact via email: Bertrand Coüasnon (bertrand.couasnon@irisa.fr), Aurélie Lemaitre (aurelie.lemaitre@irisa.fr) and Yann Soullard (yann.soullard@irisa.fr).