Architecture-Géométrie, Perception, Images, Gestes
Team manager : Isabelle SIVIGNON, Michèle ROMBAUT
The team Architecture, Geometry, Perception, Images, Gestures (AGPiG) develops algorithms for geometric modeling, image analysis and video analysis.
Research themes :
Shape reconstruction and coding. Multimedia indexing. Object segmentation and tracking. Image quality. Motion and face expression analysis. Data fusion. Parallel Programming models. Architectures and algorithms joint design exploration.
Softwares :
Real-time visual attention model
Team Presentations :
Français (pdf, 260 ko)
English (pdf, 260 ko)
A deep reinforcement learning approach for early classification of time series
Martinez Coralie, Guillaume Perrin, E Ramasso, Michèle Rombaut. A deep reinforcement learning approach for early classification of time series. 26th European Signal Processing Conference (EUSIPCO 2018), Sep 2018, Rome, Italy. 〈hal-01825472〉
Parameterization of polylines in unorganized cross-sections and open surface reconstruction
Joseph Baudrillard, Sebastien Guillon, Jean-Marc Chassery, Michèle Rombaut, Kai Wang. Parameterization of polylines in unorganized cross-sections and open surface reconstruction. 9th International Conference on Curves and Surfaces, Jun 2018, Arcachon, France. 〈hal-01774804〉
Simulation of oriented patterns with prescribed local orientation using anisotropic Gaussian fields
Kévin Polisano, Marianne Clausel, Laurent Condat, Valérie Perrier. Simulation of oriented patterns with prescribed local orientation using anisotropic Gaussian fields. [Research Report] Laboratoire Jean Kuntzmann (LJK). 2018. 〈hal-01819990〉
Voir toutes les publications de l'équipe dans HAL