
Samy LABSIR
Présentation
Samy Labsir is a associate professor at IPSA Toulouse in data science and statistics. A the same time, he carries out research works in statistical signal processing within the laboratory TéSA. He obtained its PhD Degree at University of Bordeaux (France), IMS laboratory, in the Signal&Image group with collaboration of CEA-CESTA, in December 2020. The title is "Méthodes statistiques fondées sur les groupes de Lie pour le suivi d'un amas de débris spatiaux". and was supervised by Audrey Giremus (Full professor, Bordeaux University) and Brice Yver (Engineer, CEA-CESTA). The associated manuscript is disponible here and the presentation here. From October 2020 to November 2022, he was a post-doctoral research assistant at ISAE-SUPAERO (Toulouse, France) in DEOS department, NAVIRRES research group under the supervision of Gael Pages (Researcher, ISAE-SUPAERO) and Eric Chaumette (Full professor, ISAE-SUPAERO) . Its research is focused on filtering, estimation and machine learning methods for statistical signal processing. He works on statistical (non)-bayesian methods which use the non-Euclidean properties (Lie groups and Riemannian manifolds) of both unknown parameters and observations. He also investigates on the development of (non)-Bayesian error bound for parameters evolved on (non)-Euclidean space (Lie groups).
Its research works have mainly been built for three applications:
- radar target tracking : filtering and error bound on Lie groups for parameters estimated by a radar system,
- computer vision : SLAM filtering on Lie groups from monocular camera measurements, clustering on Lie groups
- GNSS navigation : statistical studies of GNSS loop tracking, error bound on Lie groups for RTK positioning.
Compétences
Publications
Publications
|
Recurrent Neural Networks Modelling based on Riemannian Symmetric Positive Definite ManifoldEUSIPCO, Aug 2024, Lyon, France
Communication dans un congrès
hal-04708219
v1
|
|
Full Slepian-Bangs Formula for Fisher Information on Lie GroupsAsilomar Conference on Signals, Systems and Computers, 2024, Pacific Grove, United States
Communication dans un congrès
hal-04760572
v1
|
|
An Intrinsic Modified Cramér-Rao Bound on Lie Groups2024 International conference on information fusion (FUSION 2024), Jul 2024, Venise, Italy
Communication dans un congrès
hal-04620291
v1
|
Theoretical Performance Analysis of GNSS Tracking Loops2023 IEEE/ION Position, Location and Navigation Symposium (PLANS), Apr 2023, Monterey, France. pp.946-951, ⟨10.1109/PLANS53410.2023.10139980⟩
Communication dans un congrès
hal-04541103
v1
|
|
|
Intrinsic Slepian-Bangs type formula for parameters on LGs with unknown measurement noise varianceAsilomar Conference on Signals, Systems, and Computers, Oct 2023, Pacific Groove, United States
Communication dans un congrès
hal-04314563
v1
|
|
Cramér-Rao Bound on Lie Groups with Observations on Lie Groups: Application to SE(2)ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Jun 2023, Rhodes Island, Greece. ⟨10.1109/ICASSP49357.2023.10096257⟩
Communication dans un congrès
hal-04162633
v1
|
Robust M-Type Error-State Kalman Filters for Attitude Estimation2023 31st European Signal Processing Conference (EUSIPCO), Sep 2023, Helsinki, Finland. ⟨10.23919/eusipco58844.2023.10289871⟩
Communication dans un congrès
hal-04356708
v1
|
|
|
Borne de Cramér-Rao sur groupe de Lie avec observations sur groupe de Lie : application à SO(2)GRETSI 2023 - XXIXème Colloque Francophone de Traitement du Signal et des Images, Aug 2023, Grenoble, France
Communication dans un congrès
hal-04172585
v1
|
|
An Intrinsic McAulay-Seidman Bound for Parameters Evolving on Matrix Lie Groups2022 56th Asilomar Conference on Signals, Systems, and Computers, Oct 2022, Pacific Grove, France. pp.71-75, ⟨10.1109/IEEECONF56349.2022.10051899⟩
Communication dans un congrès
hal-04537127
v1
|
A Lie-group based modelling for centroid and shape estimation of a cluster of space debris2020 28th European Signal Processing Conference (EUSIPCO), Jan 2021, Amsterdam, France. pp.960-964, ⟨10.23919/Eusipco47968.2020.9287641⟩
Communication dans un congrès
hal-04538262
v1
|
|
Modèle et algorithme sur groupe de Lie pour l’estimation de la forme d’un amas de débris spatiauxColloque Gretsi, Aug 2019, Lille, France
Communication dans un congrès
hal-02531554
v1
|
|
Tracking a cluster of space debris in low orbit by filtering on Lie groupsIEEE Int. Conf. Acoust., Speech, and Signal Processing (ICASSP), May 2019, Brighton, United Kingdom
Communication dans un congrès
hal-02531539
v1
|
|
K-Means and Gaussian Mixture Models on Lie Groups: Application to Geometrical Clustering2025
Pré-publication, Document de travail
hal-05023486
v1
|
|
Méthodes statistiques fondées sur les groupes de Lie pour le suivi d'un amas de débris spatiaux.Traitement du signal et de l'image [eess.SP]. Université de Bordeaux, 2020. Français. ⟨NNT : 2020BORD0294⟩
Thèse
tel-03103892
v1
|