DYNI : DYNamiques de l’Information

Keys Words

Artificial Intelligence, Deep Learning, Cognition, Bioacoustics, Speech, Hearing, Scene Understanding, Environmental Survey, Astrophysics

 

Head

Ricard MARXER

 

Permanent Members

GLOTIN HervéProfesseur des Universités
Courriel : herve.glotin@lis-lab.fr
MARXER RicardMaitre de Conférences
Courriel : ricard.marxer@lis-lab.fr
PAIEMENT AdelineMaitre de Conférences
Courriel : adeline.paiement@lis-lab.fr
PARIS SébastienMaitre de Conférences
Courriel : sebastien.paris@lis-lab.fr
RAZIK JosephMaitre de Conférences
Courriel : joseph.razik@lis-lab.fr

 

Phd Studients

BEST PaulDoctorant
Courriel : paul.best@lis-lab.fr
FERRARI MaxenceDoctorant
Courriel : maxence.ferrari@lis-lab.fr
MORGAN JayDoctorant
Courriel : jay.morgan@lis-lab.fr
POUPARD MarionDoctorant
Courriel : marion.poupard@lis-lab.fr

 

Other staff members

BARCHASZ ValentinIE
Courriel : valentin.barchasz@lis-lab.fr
GIRAUDET PascaleCherch. associe
Courriel : pascale.giraudet@lis-lab.fr
MALIGE FranckPRAG
Courriel : franck.malige@lis-lab.fr
PATRIS JuliePRAG
Courriel : julie.patris@lis-lab.fr

 

Research interests

Our aim is develop and innovate in methods of machine learning, signal processing and data analysis in order to improve our knowledge and understanding in physical, natural and human sciences.

DYNI’s research in Artificial Intelligence and representation learning aims to cover the data acquisition, transmission and processing chain from sensors to users. The research is applied to diverse fields such as, marine and maritime robotics, bioacoustics, speech & hearing, and multimodal information analysis in physics, health or cognition. Through its technological platform SMIoT, DYNI innovates the scientific instrumentation for smart long-term data acquisition and embedded data processing.

DYNI tackles the main challenges of data-driven approaches applied to the experimental sciences: a) high cost of data acquisition, b) human bias in manual annotations, c) complexity of the underlying phenomena, and d) explicatory power of advanced machine learning models. Thus, our work focuses on unsupervised/semi-supervised learning, reinforcement learning, Deep Learning and explainable AI applied to a variety of topics ranging from environmental monitoring to the study of vocal interactivity.

Prestigious researchers are invited to DYNI and its interaction with the economic, social and cultural environment is renowned. Its activity is very intense and diversified in the field of development (software, hardware, platform). It’s pioneering research activity has led to the creation of some of the first international congresses on environmental Big Data, the coordination of the Scaled Acoustic BIODiversity project (SABIOD), the EADM action of GDR MADICS, the organisation of ICLR 2017 and the co-organisation of the VIHAR community and workshops.



 

Site Web

More detail : https://dyni.lis-lab.fr

 

Scientific publications



31 documents

Article dans une revue

  • Faegheh Sardari, Adeline Paiement, Sion Hannuna, Majid Mirmehdi. VI-Net: View-Invariant Quality of Human Movement Assessment. Sensors, MDPI, In press. ⟨hal-02934456⟩
  • Veronica Morfi, Yves Bas, Hanna Pamuła, Hervé Glotin, Dan Stowell. NIPS4Bplus: a richly annotated birdsong audio dataset. PeerJ Computer Science, PeerJ, 2019, 5, pp.e223. ⟨10.7717/peerj-cs.223⟩. ⟨hal-02321703⟩
  • Julie Patris, Franck Malige, Hervé Glotin, Mark Asch, Susannah Buchan. A standardized method of classifying pulsed sounds and its application to pulse rate measurement of blue whale southeast Pacific song units. Journal of the Acoustical Society of America, Acoustical Society of America, 2019, 146 (4), pp.2145-2154. ⟨10.1121/1.5126710⟩. ⟨hal-02440173⟩
  • Kristina Yordanova, Stefan Lüdtke, Samuel Whitehouse, Frank Krüger, Adeline Paiement, et al.. Analysing Cooking Behaviour in Home Settings: Towards Health Monitoring. Sensors, MDPI, In press. ⟨hal-02003387⟩
  • Ricard Marxer, Jon Barker, Najwa Alghamdi, Steve Maddock. The impact of the Lombard effect on audio and visual speech recognition systems. Speech Communication, Elsevier : North-Holland, 2018, 100, pp.58-68. ⟨10.1016/j.specom.2018.04.006⟩. ⟨hal-01779704⟩
  • Najwa Alghamdi, Steve Maddock, Ricard Marxer, Jon Barker, Guy Brown. A corpus of audio-visual Lombard speech with frontal and profile views. Journal of the Acoustical Society of America, Acoustical Society of America, 2018, 143 (6), pp.EL523-EL529. ⟨10.1121/1.5042758⟩. ⟨hal-01867824⟩

Communication dans un congrès

  • Joseph Jenkins, Adeline Paiement, Jean Aboudarham, Xavier Bonnin. Physics-informed detection and segmentation of type II solar radio bursts. British Machine Vision Virtual Conference, Sep 2020, Virtual, United Kingdom. ⟨hal-02923001⟩
  • Hans Dolfing, Bellegarda Jérome, Jan Chorowski, Ricard Marxer, Antoine Laurent. The "ScribbleLens" Dutch historical handwriting corpus. International Conference on Frontiers of Handwriting Recognition (ICFHR), Sep 2020, Dortmund, Germany. ⟨hal-02877520⟩
  • Paul Best, Maxence Ferrari, Marion Poupard, Sébastien Paris, Ricard Marxer, et al.. Deep Learning and Domain Transfer for Orca Vocalization Detection. International joint conference on neural networks, Jul 2020, glasgow, United Kingdom. ⟨hal-02865300⟩
  • Maxence Ferrari, Hervé Glotin, Ricard Marxer, Mark Asch. DOCC10: Open access dataset of marine mammal transient studies and end-to-end CNN classification. IJCNN, Jul 2020, Glasgow, United Kingdom. ⟨hal-02866091⟩
  • Adrian Łańcucki, Jan Chorowski, Guillaume Sanchez, Ricard Marxer, Nanxin Chen, et al.. Robust Training of Vector Quantized Bottleneck Models. IJCNN 2020, Jul 2020, Glasgow, United Kingdom. ⟨hal-02912027⟩
  • Guillaume Sanchez, Vincente Guis, Ricard Marxer, Frederic Bouchara. DEEP LEARNING CLASSIFICATION WITH NOISY LABELS. ICME Workshop, Jul 2020, Londres, United Kingdom. ⟨hal-02552375⟩
  • Marion Poupard, Benjamin de Montgolfier, Hervé Glotin. Ethoacoustic by bayesian non parametric and stochastic neighbor embedding to forecast anthropic pressure on dolphins. OCEANS, Jun 2020, Marseille, France. ⟨hal-02445440⟩
  • Alexis Joly, Hervé Goëau, Christophe Botella, Rafael Ruiz de Castaneda, Hervé Glotin, et al.. LifeCLEF 2020 Teaser: Biodiversity Identification and Prediction Challenges. 42nd European Conference on IR Research on Advances in Information Retrieval (ECIR), Apr 2020, Lisbon, Portugal. pp.542-549, ⟨10.1007/978-3-030-45442-5_70⟩. ⟨hal-02873670⟩
  • Jan Chorowski, Nanxin Chen, Ricard Marxer, Hans Dolfing, Adrian Łańcucki, et al.. Unsupervised Neural Segmentation and Clustering for Unit Discovery in Sequential Data. NeurIPS 2019 workshop - Perception as generative reasoning - Structure, Causality, Probability, Dec 2019, Vancouver, Canada. ⟨hal-02399138⟩
  • Faegheh Sardari, Adeline Paiement, Majid Mirmehdi. View-invariant Pose Analysis for Human Movement Assessment from RGB Data. 20th International Conference on Image Analysis and Processing (ICIAP), Sep 2019, Trento, Italy. ⟨10.1007/978-3-030-30645-8_22⟩. ⟨hal-02171028⟩
  • Maxence Ferrari, Ricard Marxer, Mark Asch, Hervé Glotin. Wave Propagation in the Biosonar Organ of sperm whales using a Finite Difference Time Domain method. VIHAR, Aug 2019, Lodon, United Kingdom. ⟨hal-02445408⟩
  • Marion Poupard, Paul Best, Jan Schlüter, Helena Symonds, Paul Spong, et al.. Large-scale unsupervised clustering of Orca vocalizations: a model for describing Orca communication systems. 2nd International Workshop on Vocal Interactivity in-and-between Humans, Animals and Robots, Aug 2019, Londres, United Kingdom. ⟨10.7287/peerj.preprints.27979v1⟩. ⟨hal-02965872⟩
  • Julie Patris, Dimitri Komatitsch, Maritza Sepúlveda, Macarena Santos, Hervé Glotin, et al.. Mono-hydrophone localization of baleen whales: a study of propagation using a spectral element method applied in Northern Chile. Oceans 2019, Jun 2019, Marseille, France. ⟨10.1109/OCEANSE.2019.8867333⟩. ⟨hal-02440437⟩
  • Maxence Ferrari, Marion Poupard, Pascale Giraudet, Ricard Marxer, Jean-Marc Prevot, et al.. Efficient artifacts filter by density-based clustering in long term 3D whale passive acoustic monitoring with five hydrophones fixed under an Autonomous Surface Vehicle. OCEANS 2019 - Marseille, Jun 2019, Marseille, France. pp.39, ⟨10.1109/OCEANSE.2019.8867416⟩. ⟨hal-02398121⟩
  • Maxence Ferrari, Hervé Glotin, Ricard Marxer, Valentin Barchasz, Véronique Sarano, et al.. High-frequency Near-field Physeter macrocephalus Monitoring by Stereo-Autoencoder and 3D Model of Sonar Organ. OCEANS 2019, Jun 2019, Marseille, France. ⟨hal-02313898⟩
  • Maxence Ferrari, Marion Poupard, Pascale Giraudet, Ricard Marxer, Jean-Marc Prévot, et al.. Efficient artifacts filter by density-based clustering in long term 3D whale passive acoustic monitoring with five hydrophones fixed under an Autonomous Surface Vehicle. OCEANS 2019, Jun 2019, Marseille, France. ⟨hal-02313922⟩
  • Marion Poupard, Paul Best, Jan Schlüter, Jean-Marc Prévot, Helena Symonds, et al.. Deep Learning for Ethoacoustics of Orcas on three years pentaphonic continuous recording at Orcalab revealing tide, moon and diel effects. OCEANS, Jun 2019, Marseille, France. ⟨hal-02445426⟩
  • Vincent Roger. Comment adapter les systèmes d'apprentissages modernes pour les données et problèmes en oncologie?. 1st workshop eSanté : Données, Big Data & IA en Oncologie, May 2019, Toulouse, France. ⟨hal-03003825⟩
  • M. Poupard, M. Ferrari, J Schlüter, R Marxer, P Giraudet, et al.. REAL-TIME PASSIVE ACOUSTIC 3D TRACKING OF DEEP DIVING CETACEAN BY SMALL NON-UNIFORM MOBILE SURFACE ANTENNA. 2019 IEEE International Conference on Acoustics, Speech and Signal Processing, May 2019, Brighton, United Kingdom. ⟨hal-02445414⟩
  • Jay Morgan, Adeline Paiement, Monika Seisenberger, Jane Williams, Adam Wyner. A Chatbot Framework for the Children's Legal Centre. The 31st international conference on Legal Knowledge and Information Systems (JURIX), Dec 2018, Groningen, Netherlands. ⟨hal-01878545v2⟩
  • Mandar Gogate, Ahsan Adeel, Ricard Marxer, Jon Barker, Amir Hussain. DNN Driven Speaker Independent Audio-Visual Mask Estimation for Speech Separation. Interspeech 2018, Sep 2018, Hybderabad, India. pp.2723-2727, ⟨10.21437/Interspeech.2018-2516⟩. ⟨hal-01868604⟩
  • Randall Balestriero, Romain Cosentino, Hervé Glotin, Richard Baraniuk. Spline Filters For End-to-End Deep Learning. 35th International Conference on Machine Learning, Jul 2018, stockholm, Sweden. ⟨hal-01879266⟩

Poster

  • Marion Poupard, Benjamin de Montgolfier, Vincent Roger, Devashish Lohani, Hervé Glotin. EthoAcoustics : a model based on t-SNE & Clustering, ap-plied on Pantropical spotted dolphin during Whale Watching. 8th International DCLDE (Detection, Classification, Localization, and Density Estimation) Workshop, Jun 2018, Paris, France. ⟨hal-01873336⟩

Chapitre d'ouvrage

  • Hill Kobayashi, Hiromi Kudo, Hervé Glotin, Vincent Roger, Marion Poupard, et al.. A Real-Time Streaming and Detection System for Bio-acoustic Ecological Studies after the Fukushima Accident. Multimedia Tools and Applications for Environmental & Biodiversity Informatics, 2018. ⟨hal-01879592⟩

Pré-publication, Document de travail

  • Franck Malige, Julie Patris, Susannah Buchan, Kathleen M. Stafford, Fannie Shabangu, et al.. Inter-annual decrease in pulse rate and peak frequency of Southeast Pacific blue whale song types. 2020. ⟨hal-02073928v2⟩