Institut des Systèmes Intelligents
et de Robotique

Partenariats

UPMC

CNRS

INSERM

Tremplin CARNOT Interfaces

Labex SMART

Rechercher

A voir également

Profil

NICOLLE Jérémie
Titre : Doctorant
Ne fait plus partie de l'unité

----------------------------------------------------------------------------------------------------------------------------------------------------------

Thesis

 

  • Title: Reading Faces using Hard Multi-Task Metric Learning for Kernel Regression
  • Director: Mohamed Chetouani
  • Advisor: Kevin Bailly
  • Domain: Pattern Recognition, Computer Vision

 

----------------------------------------------------------------------------------------------------------------------------------------------------------

Research

 

- Automatic Action Unit Intensity Prediction (FG'15)

  • Extending Metric Learning for Kernel Regression with a novel Multi-Task regularization

         >>>   The proposed system won the Facial Expression Recognition and Analysis challenge (FERA'15)

 

 

- Facial Landmark Localization (ICIP'13)

  • New framework for an efficient use of LBP/LPQ with binary maps
  • Triplet-based voting method for shape constraint inclusion

 

 

- Multimodal Prediction of Human Emotion in a Continuous Space (ICMI'12)

  • Including dynamics with spectral features
  • Learning temporal delay probability distributions of continuous labels

         >>>   The proposed system won the Audio/Visual Emotion Challenge (AVEC'12)

 

----------------------------------------------------------------------------------------------------------------------------------------------------------

Education

 

  • 2013 - 2015        PhD in Pattern Recognition
  •       2012              Training period (ISIR)
  • 2008 - 2012       Engineering degree (Ecole Centrale de Lyon)
  •       2012              Master's degree, Systems and Images (INSA Lyon)
  •       2010              Training periods in signal processing (IRCAM, Télécom ParisTech)
  • 2006 - 2008       Preparatory classes (Lycée Saint-Louis) Mathematics and Physics (MP*)