Deep neural networks are currently one of the most successful models in image processing and computer vision [1-4, 10-13]. Their principle consists in learning convolutional filters, together with attention [14,15] and fully connected layers, that maximize classification and generation performances. Large generative models (LGMs) [5] are a particular category of…
Socio-economic and scientific context An intracranial aneurysm (IA) is a deformation of the cerebral artery wall, leading to a fragile arterial zone that may rupture. Aneurysmal rupture most often causes a subarachnoid hemorrhage (SAH), which has a poor prognosis. It is estimated that one-third of patients with this condition will…
Un sujet de thèse est ouvert dans le cadre du projet ANR PERSAPLAN « Arthroplastie Personnalisée de l’Epaule Inversée : intégration des caractéristiques fonctionnelles et biomécaniques du patient dans le Planning chirurgical » (ANR-24-CE45-7139-01). Contexte: Le projet de thèse Perceus a pour objectif d’améliorer les résultats post-opératoires de l’arthroplastie d’épaule…
Socio-economic and scientific context Low Grade Gliomas (LGG, WHO grade II) are infiltrative brain tumours that usually occur in young patients (mean age of 38 years). Due to their slow growth and brain plasticity, most patients provide normal or nearly normal clinical examination for several years. Surgery is considered to…
General information Description Context Landmark detection, description and matching is the cornerstone of autonomous visual localization systems deployed in unknown environments. While most widely-adopted and accurate solutions exploit low-level landmarks such as points or lines, dealing with large-scale and/or visually ambiguous environments remains highly challenging due to the inherent multiplicity,…
Keywords : vision foundation models, physics-grounding, computer vision, scene understanding Supervisors : Raoul de Charette (Inria, Paris) and Tuan-Hung Vu (Inria / Valeo.ai, Paris) Description: The objective of the PhD is to improve the explicit understanding of physics in Vision Foundation Models (VFM). While the latter are typically trained on reconstruction…
The CREATIS laboratory announces the opening of a 36-month, fully funded PhD position startingin September 2025. See details in the PDF file on the top right of the page. The full description can be found here. Keywords: Image Reconstruction, Cryogenic Electron Tomography, Modeling, Deep Learning,Computational Imaging, Optimization. Summary: Cryo-electron tomography…
Laboratoire d’accueil / Host Laboratory : ImViA Laboratory, CORES team, 21078, Dijon, France Spécialité du doctorat préparé/Speciality : Instrumentation et informatique de l’image Mots-clefs / Keywords : Cryptosystem, Generative Adversarial Networks (GANs), Deep-fakes, Images, FPGA, Python, C++. Descriptif détaillé de la thèse / Job description Abstract : Detecting and safeguarding against malicious deep-fakes…
Mots-clés : Reconnaissance aveugle de modulation, systèmes de communication adaptatifs, démodulation aveugle, communication acoustique sous-marine discrète Description du projet/sujet d’étude L’identification aveugle de modulation est un domaine où convergent traitement du signal, apprentissage automatique et théorie des communications. Ce domaine vise à résoudre le problème fondamental suivant : « comment un récepteur…
