The 3D-SAM - Modeling and Analysis of Static and Dynamic Shapes team develops new models and algorithms for analysing and synthesising human movement and behaviour (bodies, faces and gestures). Our approaches are generally derived from computer vision, multimodal artificial intelligence and geometry.
Research themes:
Mohammed Daoudi
La génération et l'interaction multimodales en 3D du corps humain
Modèles Génératifs pour des Interactions naturelles entre humains et agents virtuels dotés d’une IA (GenInterHuman)
Deep Learning géométrique appliqué aux maillages
Analyse/prédiction du comportement Humain dans des séquences vidéos non contrôlées 30/09/2022
Sparse Representations in the Shape Manifold for Human Trajectories Classification and Generation. 03/12/2019
Nouvelles approches géométriques pour l'analyse du comportement humain 12/12/2018
Dynamic hand gesture recognition from traditional handcrafted to recent deep learning approaches 14/12/2017
Human object interaction recognition 09/01/2017
Semantic 3D model description – Application to example-based modeling 10/11/2016
3D human action recognition 01/12/2015
3D Dynamic Facial Sequences Analysis For Face Recognition and Emotion Detection. 02/11/2015
Analyse de formes pour la compréhension du comportement humain 02/07/2020