Research topics

The eight research topics of the department are presented below:

  • Machine learningthis is a research topic at the crossroads of several fields: applied mathematics, computer science, statistics and signal processing. The main focus of the field is to develop algorithm and methods allowing to automatically learn how to solve complexes tasks using training examples.
  • Audio data analysis and signal processing: this topic develops data analysis methods that are primarily applied to audio data. The research preformed here relies on signal processing and machine learning techniques, especially focusing on i) data decomposition et representation learning methods; ii) parametric modelling methods. The developed methodologies are used mainly for two types of tasks: i) source separation, and ii) the analysis of human activity related signals and contents, using classification techniques.
  • Computer Graphics: this research topic develops models and algorithms for 3D data, with a particular interest in geometric modeling, image synthesis, computer animation, and real-time interactive systems. The main focus is on shape, appearance, light and motion modeling in digital 3D scenes, either virtually created or captured with 3D computer vision methods, with applications in CAD, games, VFX, 3D DCC, design, virtual reality, augmented reality, digital fabrication and simulation.
  • Probability and statistics : This topic gathers research activities related to probabilistic modeling and statistical inference. These involve stochastic processes (Markov chains, time series, …), statistics in high dimension, extreme value theory and applications to anomaly detection and risk quantification, uncertainty assessment through bootstrapping and concentration inequalities for empirical processes, reinforcement learning, information theory.
  • Mathematics for images
  • Remote Sensing Data and Images : This research axis is devoted to remote sensing images and data acquired by satellite, aerial or ground based sensors. It mainly concerns Earth observation but space data are included. It focuses on the following topics: physical and mathematical modeling of acquisition systems (active and passive systems, optical and microwave imaging, Coded and Synthetic Apertures, interferometry, polarimetry,…); enhancement, information extraction, interpretation and compression of satellite or aerial images; 3D reconstruction, change and ground movement monitoring, time series; fusion of sensors.
  • Social Computing : this research topic gathers research on computational models for the analysis of social interactions, from web analysis to social robotics. The research topic is multidisciplinary: computational models are established in close collaboration with research fields such as psychology, sociology, and linguistics. They are based on methods derived from various domains in signal processing (eg speech signal processing for the recognition of emotions), in automatic learning (eg use of Conditional Random Fields for the detection of opinions in texts ), Computer science (eg automatic processing of natural language for the detection of opinions, taking into account the socio-emotional component in human-machine interactions).
  • Biomedical image, digital health: This research domain focuses on analysis and understanding of medical images, biological imaging, biomedical signal analysis, large health data bases, bio-statistics and bio-informatics, with applications in neurosciences and digital health.