Détail de l'auteur
Auteur Thierry Géraud |
Documents disponibles écrits par cet auteur (3)
Ajouter le résultat dans votre panier Affiner la recherche Interroger des sources externes
Titre : Generic programming in modern C++ for image processing Type de document : Thèse/HDR Auteurs : Michaël Roynard, Auteur ; Thierry Géraud, Directeur de thèse ; Edwin Carlinet, Directeur de thèse Editeur : Paris : Sorbonne Université Année de publication : 2022 Importance : 237 p. Format : 21 x 30 cm Note générale : bibliographie
Doctoral thesis submitted to fufill the requirements for the degree of Doctor of Sorbonne Université with the doctoral speciality of "Software Engineering and Image Processing"Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Langages informatiques
[Termes IGN] C++
[Termes IGN] langage de programmation
[Termes IGN] morphologie mathématique
[Termes IGN] programmation informatique
[Termes IGN] taxinomie
[Termes IGN] traitement d'imageIndex. décimale : THESE Thèses et HDR Résumé : (auteur) C++ is a multi-paradigm language that enables the initiated programmer to set up efficient image processing algorithms. This language strength comes from several aspects. C++ is high-level, which enables developing powerful abstractions and mixing different programming styles to ease the development. At the same time, C++ is low-level and can fully take advantage of the hardware to deliver the best performance. It is also very portable and highly compatible which allows algorithms to be called from high-level, fast-prototyping languages such as Python or Matlab. One of the most fundamental aspects where C++ really shines is generic programming. Generic programming makes it possible to develop and reuse bricks of software on objects (images) of different natures (types) without performance loss. Nevertheless,conciliating the aspects of genericity, efficiency, and simplicity is not trivial. Modern C++ (post-2011) has brought new features that made the language simpler and more powerful. In this thesis, we first explore one particular C++20aspect: the concepts, in order to build a concrete taxonomy of image related types and algorithms. Second, we explore another addition to C++20, ranges (and views), and we apply this design to image processing algorithms and image types in order to solve issues such as how hard it is to customize/tweak image processing algorithms. Finally, we explore possibilities regarding how we can offer a bridge between static (compile-time) generic C++ code and dynamic (runtime) Python code. We offer our own hybrid solution and benchmark its performance as well as discuss what can be done in the future with JIT technologies. Considering those three axes, we will address the issue regarding the way to conciliate generic programming, efficiency and ease of use. Note de contenu : I Context and History of Generic programming
1- Introduction
2- Generic programming (genericity)
II Applying Generic programming for Image processing in the static world
3- Taxonomy for Image Processing: Image types and algorithms
4- Image views
III Bringing Generic programming to the dynamic world
5- A bridge between the static world and the dynamic world
6- Conclusion and continuationNuméro de notice : 24083 Affiliation des auteurs : non IGN Thématique : IMAGERIE/INFORMATIQUE Nature : Thèse française Note de thèse : PhD thesis : Software Engineering and Image Processing : Sorbonne Université : 2022 Organisme de stage : EPITA DOI : sans En ligne : https://theses.hal.science/tel-03922670 Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=102391
Titre : ICDAR 2021 competition on historical map segmentation Type de document : Article/Communication Auteurs : Joseph Chazalon, Auteur ; Edwin Carlinet, Auteur ; Yizi Chen , Auteur ; Julien Perret
Editeur : Le Kremlin Bicêtre : Ecole pour l'Informatique et les Techniques Avancées EPITA Année de publication : 2021 Projets : 1-Pas de projet / Conférence : ICDAR 2021, 16th International Conference on Document Analysis and Recognition 05/09/2021 10/09/2021 Lausanne Suisse Importance : 15 p. Note générale : bibliographie Langues : Anglais (eng) Résumé : (auteur) This paper presents the final results of the ICDAR 2021 Competition on Historical Map Segmentation (MapSeg), encouraging research on a series of historical atlases of Paris, France, drawn at 1/5000 scale between 1894 and 1937. The competition featured three tasks, awarded separately. Task 1 consists in detecting building blocks and was won by the L3IRIS team using a DenseNet-121 network trained in a weakly supervised fashion. This task is evaluated on 3 large images containing hundreds of shapes to detect. Task 2 consists in segmenting map content from the larger map sheet, and was won by the UWB team using a U-Net-like FCN combined with a binarization method to increase detection edge accuracy. Task 3 consists in locating intersection points of geo-referencing lines, and was also won by the UWB team who used a dedicated pipeline combining binarization, line detection with Hough transform, candidate filtering, and template matching for intersection refinement. Tasks 2 and 3 are evaluated on 95 map sheets with complex content. Dataset, evaluation tools and results are available under permissive licensing at https://icdar21-mapseg.github.io/. Numéro de notice : C2021-022 Affiliation des auteurs : UGE-LASTIG+Ext (2020- ) Nature : Communication nature-HAL : ComAvecCL&ActesPubliésIntl DOI : sans En ligne : https://hal.science/hal-03256193 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=98032
Titre : Introducing the boundary-aware loss for deep image segmentation Type de document : Article/Communication Auteurs : Minh On Vu Ngoc, Auteur ; Yizi Chen Editeur : The British Machine Vision Association Press (BMVA) Année de publication : 2021 Projets : SODUCO / Perret, Julien Conférence : BMVC 2021, 32nd British Machine Vision Conference 22/11/2021 25/11/2021 online Royaume-Uni OA Proceedings Importance : 17 p. Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image
[Termes IGN] apprentissage profond
[Termes IGN] classification barycentrique
[Termes IGN] segmentation d'imageRésumé : (auteur) Most contemporary supervised image segmentation methods do not preserve the initial topology of the given input (like the closeness of the contours). One can generally remark that edge points have been inserted or removed when the binary prediction and the ground truth are compared. This can be critical when accurate localization of multiple interconnected objects is required. In this paper, we present a new loss function, called, Boundary-Aware loss (BALoss), based on the Minimum Barrier Distance (MBD) cut algorithm. It is able to locate what we call the leakage pixels and to encode the boundary information coming from the given ground truth. Thanks to this adapted loss, we are able to significantly refine the quality of the predicted boundaries during the learning procedure. Furthermore, our loss function is differentiable and can be applied to any kind of neural network used in image processing. We apply this loss function on the standard U-Net and DC U-Net on Electron Microscopy datasets. They are well-known to be challenging due to their high noise level and to the close or even connected objects covering the image space. Our segmentation performance, in terms of Variation of Information (VOI) and Adapted Rank Index (ARI), are very promising and lead to 15% better scores of VOI and 5% better scores of ARI than the state-of-the-art. The code of boundary-awareness loss is freely available at https://github.com/onvungocminh/MBD_BAL Numéro de notice : C2021-054 Affiliation des auteurs : UGE-LASTIG+Ext (2020- ) Thématique : IMAGERIE Nature : Communication nature-HAL : ComAvecCL&ActesPubliésIntl DOI : sans En ligne : https://www.bmvc2021-virtualconference.com/assets/papers/1546.pdf Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99411
