Search

Your search keyword '"Marin Ferecatu"' showing total 26 results
Start Over Author "Marin Ferecatu" Publication Year Range Last 10 years
26 results on '"Marin Ferecatu"'

1. Processing of Extremely High-Resolution LiDAR and RGB Data: Outcome of the 2015 IEEE GRSS Data Fusion Contest–Part A: 2-D Contest

Author

Manuel Campos-Taberner, Adriana Romero-Soriano, Carlo Gatta, Gustau Camps-Valls, Adrien Lagrange, Bertrand Le Saux, Anne Beaupere, Alexandre Boulch, Adrien Chan-Hon-Tong, Stephane Herbin, Hicham Randrianarivo, Marin Ferecatu, Michal Shimoni, Gabriele Moser, and Devis Tuia

Subjects
Deep neural networks, extremely high spatial resolution, image analysis and data fusion (IADF), landcover classification, LiDAR, multiresolution-, Ocean engineering, TC1501-1800, Geophysics. Cosmic physics, QC801-809
Abstract

In this paper, we discuss the scientific outcomes of the 2015 data fusion contest organized by the Image Analysis and Data Fusion Technical Committee (IADF TC) of the IEEE Geoscience and Remote Sensing Society (IEEE GRSS). As for previous years, the IADF TC organized a data fusion contest aiming at fostering new ideas and solutions for multisource studies. The 2015 edition of the contest proposed a multiresolution and multisensorial challenge involving extremely high-resolution RGB images and a three-dimensional (3-D) LiDAR point cloud. The competition was framed in two parallel tracks, considering 2-D and 3-D products, respectively. In this paper, we discuss the scientific results obtained by the winners of the 2-D contest, which studied either the complementarity of RGB and LiDAR with deep neural networks (winning team) or provided a comprehensive benchmarking evaluation of new classification strategies for extremely high-resolution multimodal data (runner-up team). The data and the previously undisclosed ground truth will remain available for the community and can be obtained at http://www.grss-ieee.org/community/technical-committees/data-fusion/2015-ieee-grss-data-fusion-contest/. The 3-D part of the contest is discussed in the Part-B paper [1].

Published
2016
Full Text
View/download PDF

18. Deep Online Storage-Free Learning on Unordered Image Streams

Author

Pierre Blanchart, Marin Ferecatu, Michel Crucianu, Andrey Besedin, Intelligence Artificielle et Apprentissage Automatique (LI3A), Département Métrologie Instrumentation & Information (DM2I), Laboratoire d'Intégration des Systèmes et des Technologies (LIST (CEA)), Direction de Recherche Technologique (CEA) (DRT (CEA)), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Direction de Recherche Technologique (CEA) (DRT (CEA)), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Université Paris-Saclay-Laboratoire d'Intégration des Systèmes et des Technologies (LIST (CEA)), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Université Paris-Saclay, CEDRIC. Données complexes, apprentissage et représentations (CEDRIC - VERTIGO), Centre d'études et de recherche en informatique et communications (CEDRIC), Ecole Nationale Supérieure d'Informatique pour l'Industrie et l'Entreprise (ENSIIE)-Conservatoire National des Arts et Métiers [CNAM] (CNAM), HESAM Université - Communauté d'universités et d'établissements Hautes écoles Sorbonne Arts et métiers université (HESAM)-HESAM Université - Communauté d'universités et d'établissements Hautes écoles Sorbonne Arts et métiers université (HESAM)-Ecole Nationale Supérieure d'Informatique pour l'Industrie et l'Entreprise (ENSIIE)-Conservatoire National des Arts et Métiers [CNAM] (CNAM), HESAM Université - Communauté d'universités et d'établissements Hautes écoles Sorbonne Arts et métiers université (HESAM)-HESAM Université - Communauté d'universités et d'établissements Hautes écoles Sorbonne Arts et métiers université (HESAM), Springer, Anna Monreale, Carlos Alzate, Michael Kamp, Yamuna Krishnamurthy, Daniel Paurat, Moamar Sayed-Mouchaweh, Albert Bifet, João Gama, Rita P. Ribeiro, Laboratoire d'analyse des données et d'intelligence des systèmes (LADIS), Laboratoire d'Intégration des Systèmes et des Technologies (LIST), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Université Paris-Saclay-Laboratoire d'Intégration des Systèmes et des Technologies (LIST), and Ecole Nationale Supérieure d'Informatique pour l'Industrie et l'Entreprise (ENSIIE)-Conservatoire National des Arts et Métiers [CNAM] (CNAM)-Ecole Nationale Supérieure d'Informatique pour l'Industrie et l'Entreprise (ENSIIE)-Conservatoire National des Arts et Métiers [CNAM] (CNAM)

Subjects
Data stream, Computer science, online learning, 02 engineering and technology, 010501 environmental sciences, Reuse, [INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE], Machine learning, computer.software_genre, 01 natural sciences, Image (mathematics), [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], 0202 electrical engineering, electronic engineering, information engineering, 0105 earth and related environmental sciences, Generative Adversarial Networks, Forgetting, business.industry, Data stream mining, Deep learning, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], artificial intelligence, Classification, GAN, 020201 artificial intelligence & image processing, Artificial intelligence, business, computer, MNIST database, Generative grammar, Data streams
Abstract

International audience; In this work we develop an online deep-learning based approach for classification on data streams. Our approach is able to learn in an incremental way without storing and reusing the historical data (we only store a recent history) while processing each new data sample only once. To make up for the absence of the historical data, we train Generative Adversarial Networks (GANs), which, in recent years have shown their excellent capacity to learn data distributions for image datasets. We test our approach on MNIST and LSUN datasets and demonstrate its ability to adapt to previously unseen data classes or new instances of previously seen classes, while avoiding forgetting of previously learned classes/instances of classes that do not appear anymore in the data stream.

Published
2018
Full Text
View/download PDF

19. On the Beneficial Effect of Noise in Vertex Localization

Author

Dionysios D. Sourlas, Stefanos Kollias, Marin Ferecatu, Konstantinos A. Raftopoulos, Image, Video and Multimedia Laboratory (IVML), National Technical University of Athens [Athens] (NTUA), CEDRIC. Données complexes, apprentissage et représentations (CEDRIC - VERTIGO), Centre d'études et de recherche en informatique et communications (CEDRIC), Ecole Nationale Supérieure d'Informatique pour l'Industrie et l'Entreprise (ENSIIE)-Conservatoire National des Arts et Métiers [CNAM] (CNAM)-Ecole Nationale Supérieure d'Informatique pour l'Industrie et l'Entreprise (ENSIIE)-Conservatoire National des Arts et Métiers [CNAM] (CNAM), University of Lincoln, and Unilever R&D

Subjects
G740 Computer Vision, Perspective (graphical), Process (computing), Boundary (topology), [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], 020207 software engineering, 02 engineering and technology, Topology, Curvature, Vertex (geometry), Identification (information), Noise, Artificial Intelligence, [INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV], Pattern recognition (psychology), 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, Computer Vision and Pattern Recognition, Software, Mathematics
Abstract

International audience; A theoretical and experimental analysis related to the effect of noise in the task of vertex identification in unknown shapes is presented. Shapes are seen as real functions of their closed boundary. An alternative global perspective of curvature is examined providing insight into the process of noise-enabled vertex localization. The analysis reveals that noise facilitates in the localization of certain vertices. The concept of noising is thus considered and a relevant global method for localizing Global Vertices is investigated in relation to local methods under the presence of increasing noise. Theoretical analysis reveals that induced noise can indeed help localizing certain vertices if combined with global descriptors. Experiments with noise and a comparison to localized methods validate the theoretical results.

Published
2018
Full Text
View/download PDF

20. Deep learning for urban remote sensing

Author

Hicham Randrianarivo, Renaud Marlet, Nicolas Audebert, Alexandre Boulch, Marin Ferecatu, Sébastien Lefèvre, Bertrand Le Saux, Environment observation with complex imagery (OBELIX), Université de Bretagne Sud (UBS)-SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE (IRISA-D5), Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Université de Bretagne Sud (UBS)-École normale supérieure - Rennes (ENS Rennes)-Institut National de Recherche en Informatique et en Automatique (Inria)-Télécom Bretagne-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Université de Rennes 1 (UR1), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Université de Bretagne Sud (UBS)-École normale supérieure - Rennes (ENS Rennes)-Institut National de Recherche en Informatique et en Automatique (Inria)-Télécom Bretagne-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-École normale supérieure - Rennes (ENS Rennes)-Institut National de Recherche en Informatique et en Automatique (Inria)-Télécom Bretagne-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), ONERA - The French Aerospace Lab [Palaiseau], ONERA-Université Paris Saclay (COmUE), Centre d'études et de recherche en informatique et communications (CEDRIC), Ecole Nationale Supérieure d'Informatique pour l'Industrie et l'Entreprise (ENSIIE)-Conservatoire National des Arts et Métiers [CNAM] (CNAM), HESAM Université - Communauté d'universités et d'établissements Hautes écoles Sorbonne Arts et métiers université (HESAM)-HESAM Université - Communauté d'universités et d'établissements Hautes écoles Sorbonne Arts et métiers université (HESAM), Laboratoire d'Informatique Gaspard-Monge (LIGM), Université Paris-Est Marne-la-Vallée (UPEM)-École des Ponts ParisTech (ENPC)-ESIEE Paris-Fédération de Recherche Bézout-Centre National de la Recherche Scientifique (CNRS), imagine [Marne-la-Vallée], Université Paris-Est Marne-la-Vallée (UPEM)-École des Ponts ParisTech (ENPC)-ESIEE Paris-Fédération de Recherche Bézout-Centre National de la Recherche Scientifique (CNRS)-Université Paris-Est Marne-la-Vallée (UPEM)-École des Ponts ParisTech (ENPC)-ESIEE Paris-Fédération de Recherche Bézout-Centre National de la Recherche Scientifique (CNRS), CentraleSupélec-Télécom Bretagne-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National de Recherche en Informatique et en Automatique (Inria)-École normale supérieure - Rennes (ENS Rennes)-Université de Bretagne Sud (UBS)-Centre National de la Recherche Scientifique (CNRS)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Télécom Bretagne-Université de Rennes 1 (UR1), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National de Recherche en Informatique et en Automatique (Inria)-École normale supérieure - Rennes (ENS Rennes)-Centre National de la Recherche Scientifique (CNRS)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA), Centre National de la Recherche Scientifique (CNRS)-Fédération de Recherche Bézout-ESIEE Paris-École des Ponts ParisTech (ENPC)-Université Paris-Est Marne-la-Vallée (UPEM), Centre National de la Recherche Scientifique (CNRS)-Fédération de Recherche Bézout-ESIEE Paris-École des Ponts ParisTech (ENPC)-Université Paris-Est Marne-la-Vallée (UPEM)-Centre National de la Recherche Scientifique (CNRS)-Fédération de Recherche Bézout-ESIEE Paris-École des Ponts ParisTech (ENPC)-Université Paris-Est Marne-la-Vallée (UPEM), SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE (IRISA-D5), Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Université de Bretagne Sud (UBS)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure - Rennes (ENS Rennes)-Institut National de Recherche en Informatique et en Automatique (Inria)-Télécom Bretagne-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-CentraleSupélec-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Université de Rennes (UNIV-RENNES)-CentraleSupélec-Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Université de Rennes (UNIV-RENNES)-CentraleSupélec, ONERA, Laboratoire d'Informatique Gaspard-Monge (ligm), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Université de Bretagne Sud (UBS)-École normale supérieure - Rennes (ENS Rennes)-Institut National de Recherche en Informatique et en Automatique (Inria)-Télécom Bretagne-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Université de Bretagne Sud (UBS)-École normale supérieure - Rennes (ENS Rennes)-Institut National de Recherche en Informatique et en Automatique (Inria)-Télécom Bretagne-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), and Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-École normale supérieure - Rennes (ENS Rennes)-Institut National de Recherche en Informatique et en Automatique (Inria)-Télécom Bretagne-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)

Subjects
Earth observation, Computer science, Feature extraction, 0211 other engineering and technologies, Context (language use), 02 engineering and technology, Hough transform, law.invention, Labeling, law, 0202 electrical engineering, electronic engineering, information engineering, Training, Computer vision, 021101 geological & geomatics engineering, Remote sensing, Support vector machines, Contextual image classification, business.industry, Hough transforms, Deep learning, 020207 software engineering, Semantics, Support vector machine, Semantic mapping, [INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV], Three-dimensional displays, learning (artificial intelligence), Artificial intelligence, business, image classification
Abstract

International audience; This work shows how deep learning techniques can benefit to remote sensing. We focus on tasks which are recurrent in Earth Observation data analysis. For classification and semantic mapping of aerial images, we present various deep network architectures and show that context information and dense labeling allow to reach better performances. For estimation of normals in point clouds, combining Hough transform with convolutional networks also improves the accuracy of previous frameworks by detecting hard configurations like corners. It shows that deep learning allows to revisit remote sensing and offers promising paths for urban modeling and monitoring.

Published
2017
Full Text
View/download PDF

21. A Global-Local Approach to Extracting Deformable Fashion Items from Web Images

Author

Michel Crucianu, Helena Rodriguez, Marin Ferecatu, Lixuan Yang, CEDRIC. Données complexes, apprentissage et représentations (CEDRIC - VERTIGO), Centre d'études et de recherche en informatique et communications (CEDRIC), Ecole Nationale Supérieure d'Informatique pour l'Industrie et l'Entreprise (ENSIIE)-Conservatoire National des Arts et Métiers [CNAM] (CNAM)-Ecole Nationale Supérieure d'Informatique pour l'Industrie et l'Entreprise (ENSIIE)-Conservatoire National des Arts et Métiers [CNAM] (CNAM), Auteur indépendant, CIFRE, Enqing Chen, Yihong Gong, and Yun Tie

Subjects
Active Contour, Computer science, 02 engineering and technology, 010501 environmental sciences, computer.software_genre, 01 natural sciences, Set (abstract data type), Segmentation, Component (UML), 0202 electrical engineering, electronic engineering, information engineering, Local search (optimization), 0105 earth and related environmental sciences, Active contour model, Parsing, business.industry, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], Pattern recognition, Object (computer science), Clothing extraction, GrabCut, 020201 artificial intelligence & image processing, Artificial intelligence, business, computer
Abstract

Proceedings of the Pacific-Rim Conference on Multimedia, Xi´an, China, September 15-16, 2016, Part II; International audience; In this work we propose a new framework for extracting deformable clothing items from images by using a three stage global-local fitting procedure. First, a set of initial segmentation templates are generated from a handcrafted database. Then, each template initiates an object extraction process by a global alignment of the model, followed by a local search minimizing a measure of the misfit with respect to the potential boundaries in the neighborhood. Finally, the results provided by each template are aggregated, with a global fitting criterion, to obtain the final segmentation. The method is validated on the Fashionista database and on a new database of manually segmented images. Our method compares favorably with the Paper Doll clothing parsing and with the recent GrabCut on One Cut foreground extraction method. We quantitatively analyze each component, and show examples of both successful segmentation and difficult cases.

Published
2016
Full Text
View/download PDF

22. Traitement de données RGB et Lidar à extrêmement haute résolution: retombées de la compétition de fusion de données 2015 de l'IEEE GRSS - Partie A / compétition 2D

Author

Devis Tuia, Bertrand Le Saux, Hicham Randrianarivo, Adriana Romero-Soriano, Marin Ferecatu, Adrien Lagrange, Stéphane Herbin, Anne Beaupere, Gabriele Moser, Adrien Chan-Hon-Tong, Michal Shimoni, Gustau Camps-Valls, Alexandre Boulch, Carlo Gatta, Manuel Campos-Taberner, Universitat de València (UV), Universitat de Barcelona (UB), Universitat Autònoma de Barcelona (UAB), ONERA - The French Aerospace Lab [Palaiseau], ONERA-Université Paris Saclay (COmUE), École Nationale Supérieure de Techniques Avancées (ENSTA Paris), Centre d'études et de recherche en informatique et communications (CEDRIC), Ecole Nationale Supérieure d'Informatique pour l'Industrie et l'Entreprise (ENSIIE)-Conservatoire National des Arts et Métiers [CNAM] (CNAM), Royal Military Academy (RMA), University of Genoa (UNIGE), Universität Zürich [Zürich] = University of Zurich (UZH), Centre National de la Recherche Scientifique - CNRS (FRANCE), Institut National Polytechnique de Toulouse - INPT (FRANCE), Office National d'Etudes et Recherches Aérospatiales - ONERA (FRANCE), Université Toulouse III - Paul Sabatier - UT3 (FRANCE), Université Toulouse - Jean Jaurès - UT2J (FRANCE), Université Toulouse 1 Capitole - UT1 (FRANCE), Institut National Polytechnique de Toulouse - Toulouse INP (FRANCE), University of Zurich, and Tuia, Devis

Subjects
Atmospheric Science, 010504 meteorology & atmospheric sciences, Computer science, MULTIMODAL-DATA FUSION, Geophysics. Cosmic physics, 0211 other engineering and technologies, 02 engineering and technology, CONTEST, computer.software_genre, 01 natural sciences, Outcome (game theory), LIDAR, Traitement des images, IMAGE ANALYSIS AND DATA FUSION (IADF), DEEP NEURAL NETWORKS, Deep neural networks, Traitement du signal et de l'image, MULTIRESOLUTION, 910 Geography & travel, Multiresolution, Ground truth, LANDCOVER CLASSIFICATION, IMAGE AERIENNE, 1903 Computers in Earth Sciences, Benchmarking, Vision par ordinateur et reconnaissance de formes, Ocean engineering, 10122 Institute of Geography, Lidar, Data mining, Extremely high spatial resolution, Multimodal-data fusion, LiDAR, Computers in Earth Sciences, Image analysis and data fusion (IADF), EXTREMELY HIGH SPATIAL RESOLUTION, CLASSIFICATION, TRAITEMENT IMAGE, 1902 Atmospheric Science, APPRENTISSAGE STATISTIQUE, TELEDETECTION, Synthèse d'image et réalité virtuelle, TC1501-1800, 021101 geological & geomatics engineering, 0105 earth and related environmental sciences, Landcover classification, multiresolution, [INFO.INFO-DB]Computer Science [cs]/Databases [cs.DB], QC801-809, Intelligence artificielle, MULTISOURCE, Sensor fusion, RGB color model, computer, Multisource
Abstract

International audience; In this paper, we discuss the scientific outcomes of the 2015 data fusion contest organized by the Image Analysis and Data Fusion Technical Committee (IADF TC) of the IEEE Geoscience and Remote Sensing Society (IEEE GRSS). As for previous years, the IADF TC organized a data fusion contest aiming at fostering new ideas and solutions for multisource studies. The 2015 edition of the contest proposed a multiresolution and multisensorial challenge involving extremely high-resolution RGB images and a three-dimensional (3-D) LiDAR point cloud. The competition was framed in two parallel tracks, considering 2-D and 3-D products, respectively. In this paper, we discuss the scientific results obtained by the winners of the 2-D contest, which studied either the complementarity of RGB and LiDAR with deep neural networks (winning team) or provided a comprehensive benchmarking evaluation of new classification strategies for extremely high-resolution multimodal data (runner-up team). The data and the previously undisclosed ground truth will remain available for the community and can be obtained at http://www.grss-ieee.org/community/technical-committees/data-fusion/2015-ieee-grss-data-fusion-contest/. The 3-D part of the contest is discussed in the Part-B paper [1].

Published
2016
Full Text
View/download PDF

23. Fast Action Localization in Large-Scale Video Archives

Author

Michel Crucianu, Andrei Stoian, Marin Ferecatu, Jenny Benois-Pineau, CEDRIC. Données complexes, apprentissage et représentations (CEDRIC - VERTIGO), Centre d'études et de recherche en informatique et communications (CEDRIC), Ecole Nationale Supérieure d'Informatique pour l'Industrie et l'Entreprise (ENSIIE)-Conservatoire National des Arts et Métiers [CNAM] (CNAM)-Ecole Nationale Supérieure d'Informatique pour l'Industrie et l'Entreprise (ENSIIE)-Conservatoire National des Arts et Métiers [CNAM] (CNAM), Laboratoire Bordelais de Recherche en Informatique (LaBRI), and Université de Bordeaux (UB)-Centre National de la Recherche Scientifique (CNRS)-École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB)

Subjects
Computer science, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Feature selection, 02 engineering and technology, global alignment (GA), Machine learning, computer.software_genre, [SPI]Engineering Sciences [physics], feature selection, Discriminative model, Histogram, 0202 electrical engineering, electronic engineering, information engineering, Media Technology, [INFO]Computer Science [cs], Electrical and Electronic Engineering, scalability, Information retrieval, business.industry, Frame (networking), [INFO.INFO-MM]Computer Science [cs]/Multimedia [cs.MM], 020207 software engineering, Visualization, cascade, Support vector machine, Filter (video), Video tracking, 020201 artificial intelligence & image processing, Artificial intelligence, business, computer, Action localization
Abstract

International audience; Finding content in large video archives has so far required textual annotation to enable search by keywords. Our aim is to support retrieval from such archives using queries based on the example video clips that contain meaningful human actions. We propose a solution for the scalable search of actions in large-scale archives by leveraging the complementarity between the description at the frame level and the aggregation in time of descriptors. To permit fast search, we introduce a two-level cascade. The inexpensive first level employs aggregation to filter out a large part of the video. At the second level, aided by feature selection, a more discriminative comparison by frame alignment ranks the remaining video sequences. We improve upon the state of the art on popular data sets, and we introduce and show the results on a novel video archive data set that is significantly larger than previous ones.

Published
2015
Full Text
View/download PDF

24. Scalable action localization with kernel-space hashing

Author

Jenny Benois-Pineau, Marin Ferecatu, Michel Crucianu, Andrei Stoian, CEDRIC. Données complexes, apprentissage et représentations (CEDRIC - VERTIGO), Centre d'études et de recherche en informatique et communications (CEDRIC), Ecole Nationale Supérieure d'Informatique pour l'Industrie et l'Entreprise (ENSIIE)-Conservatoire National des Arts et Métiers [CNAM] (CNAM)-Ecole Nationale Supérieure d'Informatique pour l'Industrie et l'Entreprise (ENSIIE)-Conservatoire National des Arts et Métiers [CNAM] (CNAM), Laboratoire Bordelais de Recherche en Informatique (LaBRI), and Université de Bordeaux (UB)-Centre National de la Recherche Scientifique (CNRS)-École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB)

Subjects
Sublinear function, business.industry, hyperplane queries, Hash function, Detector, kernel indexing, [INFO.INFO-MM]Computer Science [cs]/Multimedia [cs.MM], Pattern recognition, 0102 computer and information sciences, 02 engineering and technology, 01 natural sciences, [SPI]Engineering Sciences [physics], Kernel (image processing), 010201 computation theory & mathematics, 020204 information systems, Scalability, 0202 electrical engineering, electronic engineering, information engineering, query-by-detector, [INFO]Computer Science [cs], Artificial intelligence, LSH, business, Action localization, Mathematics, Linear search
Abstract

International audience; To detect and locate complex human actions in video, one trains a detector for each target class and applies it to the video content. This approach can scale to large video databases if the application of the detector can be made sublinear in the size of the database. Sublinear retrieval methods have been successfully explored for query-by-example but few were devised for these more challenging queries by detector. We put forward here a novel approximate search method that relies on LSH to support query-by-detector. We evaluate our method on a recent large action localization dataset and show it has significantly better efficiency than linear search.

Published
2015
Full Text
View/download PDF

25. Local integrity constraints for structure detection and segmentation in high-resolution earth observation images

Author

Marin Ferecatu and Pierre Blanchart

Subjects
Earth observation, Computer science, Segmentation-based object categorization, business.industry, Scale-space segmentation, Pattern recognition, Image segmentation, Image texture, Minimum spanning tree-based segmentation, Region growing, Data integrity, Segmentation, Computer vision, Artificial intelligence, business
Abstract

Considering the idea that objects in images have a higher local structural integrity than the background they lie into, we propose a method that learns a supervised distance characterizing the membership of a pair of elements to the target structure. We test our ideas by applying them to the task of extracting semantic structures in high resolution Earth Observation images. The results show that the method works well in many situations when there is no training dataset. The limits of the method are also discussed.1

Published
2015
Full Text
View/download PDF

26. Benchmarking classification of earth-observation data: From learning explicit features to convolutional networks

Author

Anne Beaupere, Adrien Lagrange, Alexandre Boulch, Adrien Chan-Hon-Tong, Marin Ferecatu, Hicham Randrianarivo, Bertrand Le Saux, and Stéphane Herbin

Subjects
Earth observation, Artificial neural network, Computer science, business.industry, Deep learning, Feature extraction, Pattern recognition, Machine learning, computer.software_genre, Support vector machine, Benchmark (computing), Artificial intelligence, business, Transfer of learning, computer
Abstract

In this paper, we address the task of semantic labeling of multisource earth-observation (EO) data. Precisely, we benchmark several concurrent methods of the last 15 years, from expert classifiers, spectral support-vector classification and high-level features to deep neural networks. We establish that (1) combining multisensor features is essential for retrieving some specific classes, (2) in the image domain, deep convolutional networks obtain significantly better overall performances and (3) transfer of learning from large generic-purpose image sets is highly effective to build EO data classifiers.

Published
2015
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results