Visitor
Show Details - Back to research page - Back to home page
Université Pierre et Marie Curie - Paris VI - December 15, 2004
Journal of Real-Time Image Processing (jrtip23), Volume 20, Number 6, 2023
(PR23), Pattern Recognition, Volume 136, Number 109207, 2023
Computer Vision and Image Understanding (CVIU20), Volume 197-198, 2020
International Journal an Document Analysis and Recognition (IJDAR16), Volume 19, 2016, Pages 99–117
Image and Vision Computing (IVC16), Volume 46, 2016
Pattern Analysis and Applications (PAA13), Volume 16, Issue 4, nov 2013, Pages 519–533
Signal Processing: Image Communication (SPIC12), Volume 27, Issue 2, February 2012, Pages 153–171
Pattern Recognition Letters (PRL09), Volume: 30, Issue: 6, April 2009.
International Journal of Image and Graphics (IJIG08), Volume: 8, Issue: 1 (January 2008)
Revue Traitement du Signal (TS05), Special Issue on Omnidirectional Vision, Volume 22, Number 5, pages 527-535, December 2005
13th IAPR International Workshop on Document Analysis Systems (DAS18)
13th IAPR International Workshop on Document Analysis Systems (DAS18)
International Conference on Computer Vision Theory and Applications (VISAPP16)
Proceedings of the 13th IAPR International Conference on Document Analysis and Recognition (ICDAR15)
VISAPP2015
VISAPP2015
IEEE International Conference on Image Processing (ICIP14). Paris, France. Oct 27-30 2014.
IEEE International Conference on Image Processing (ICIP 10). September 26-29, 2010. Hong Kong.
IEEE International Conference on Image Processing (ICIP09). November 7-10, 2009. Cairo, Egypt.
City Models, Roads and Traffic (ISPRS Workshop - CMRT09). September 3-4, 2009. Paris, France
International Symposium on Mathematical Morphology (ISMM'09). Groningen, The Netherlands, August 24-27, 2009.
31th meeting of the International Society for Stereology (ISS08), French section, Paris, Ecole des Mines, 7 february 2008
COmpression et REprésentation des Signaux Audiovisuels (CORESA'07). Montpellier, France, Nov. 8-9 2007.
International Conference of Image Processing (ICIP'07). San Antonio (Texas), USA, Oct. 16-19 2007.
International Conference on Computer Vision and Graphics (ICCVG'04), warsaw, september 2004
Published by Kluwer in the book series COMPUTATIONAL IMAGING AND VISION.
19th GRETSI Symposium on Signal and Image Processing (GRETSI'03), Paris, 8 - 11 September 2003.
IEEE European Conference Computer Vision, Workshop on Omnidirectional Vision (ECCV Workshop-OMNIVIS 02), page 45-52, Copenhague, 02 Juin 2002.
Séminaire Perfromance et Généricité EPITA Research and Development Laboratory, Le Kremlin-Bicêtre, France, 18 mai 2011
"L'algorithme « ICP »", J. Fabrizio
Séminaire mouvement, Centre de Morphologie Mathematique, Ecole des Mines de Paris, Fontainebleau, 19 mars 2008.
"Détection et localisation d'obstacles et apports des capteurs catadioptriques pour la navigation", J. Fabrizio, O. Kloc, J. Devars,
Journées de synthèse ARCOS, Arcueil 25 Septembre 2003
"Détection d'obstacles coopératifs", J. Fabrizio, O. Kloc, J. Devars,
Journées de synthè, Satory, 19 juin 2003
"Solution analytique du « PNP » étendue aux capteurs cartadioptriques", J. Fabrizio, J. Devars,
Journée Vision Panoramique, GT5 Vision - GDR ISIS, Paris, 5 juin 2003.
Technical document 06/20
Mes publications/My publications
PhD Thesis
[+] "Localisation d'obstacles coopératifs par systèmes de vision classiques et panoramiques", J. FabrizioUniversité Pierre et Marie Curie - Paris VI - December 15, 2004
Resume
Le bilan humain et financier des accidents routiers est lourd, et l'erreur humaine est souvent en cause. Le projet ARCOS, piloté par l'INREST, est dédié à la recherche de solutions. Dans le cadre de ce projet, nous proposons un système de détection d'obstacles basé sur une approche coopérative de la vision. Les obstacles (véhicules environnants entre autres) sont balisés au moyen d'un marquage visible uniquement dans le proche infrarouge. Ce marquage permet la détection ainsi que la localisation 3D d'un obstacle à l'avant du véhicule à l'aide d'une caméra classique. Nous proposons ensuite l'ajout d'un capteur catadioptrique panoramique permettant de surveiller l'intégralité de l'environnement. Combiné avec la caméra, une paire stéréo est ainsi obtenue, venant renforcer le système coopératif. Des solutions sont donc proposées pour généraliser la technique de localisation aux capteurs catadioptriques, calibrer ce type de capteurs, et estimer la géométrie épipolaire de paires stéréo obtenues à partir de couples hétérogènes de capteurs.
Abstract
...
Thesis Advisor
Jean DEVARS
Jury
Citations : 2
Bibtex
Le bilan humain et financier des accidents routiers est lourd, et l'erreur humaine est souvent en cause. Le projet ARCOS, piloté par l'INREST, est dédié à la recherche de solutions. Dans le cadre de ce projet, nous proposons un système de détection d'obstacles basé sur une approche coopérative de la vision. Les obstacles (véhicules environnants entre autres) sont balisés au moyen d'un marquage visible uniquement dans le proche infrarouge. Ce marquage permet la détection ainsi que la localisation 3D d'un obstacle à l'avant du véhicule à l'aide d'une caméra classique. Nous proposons ensuite l'ajout d'un capteur catadioptrique panoramique permettant de surveiller l'intégralité de l'environnement. Combiné avec la caméra, une paire stéréo est ainsi obtenue, venant renforcer le système coopératif. Des solutions sont donc proposées pour généraliser la technique de localisation aux capteurs catadioptriques, calibrer ce type de capteurs, et estimer la géométrie épipolaire de paires stéréo obtenues à partir de couples hétérogènes de capteurs.
Abstract
...
Thesis Advisor
Jean DEVARS
Jury
Mme Danièle FOURNIER | Présidente | ESPCI | UPMC |
M. Jean SEQUEIRA | Rapporteur | LSIS | ESIL |
M. El Mustapha MOUADDIB | Rapporteur | CREA | UPJV |
M. Peter STURM | Examinateur | MOVI | INRIA |
M. Maurice MILGRAM | Examinateur | LISIF | UPMC |
M. Jean DEVARS | Examinateur | LISIF | UPMC |
Citations : 2
- "Contributions à la vision omnidirectionnelle: Étude, Conception et Étalonnage de capteurs pour l'acquisition d'images et la modélisation 3D", Bertrand VANDEPORTAELE (PhD Thesis 2006)
- "Etude des proprietes geometriques des capteurs panoramiques", Sio-Hoi IENG (PhD Thesis 2005)*
Bibtex
@PhdThesis{fabrizio-phd04, author = "Jonathan Fabrizio", title = "Localisation d'obstacles coopératifs par systèmes de vision classiques et panoramiques" school = "Université Pierre et Marie Curie - Paris VI", year = "2004", month = "decembre", url = {http://jo.fabrizio.free.fr/recherche/th_fabrizio.pdf} }
Journal articles
[+] "How to compute the convex hull of a binary shape? A real-time algorithm to compute the convex hull of a binary shape", Jonathan FabrizioJournal of Real-Time Image Processing (jrtip23), Volume 20, Number 6, 2023
Abstract :
In this article, we present an algorithm to compute the convex hull of a binary shape. Efficient algorithms to compute the convex hull of a set of points had been proposed long time ago. For a binary shape, the common practice is to rely on one of them: to compute the convex hull of binary shape, all pixels of the shape are first listed, and then the convex hull is computed on this list of points. The computed convex hull is finally rasterized to provide the final result (as, for example, in the famous scikit-image library). To compute the convex hull of an arbitrary set of points, the points of the list that lie on the outline of the convex hull must be selected (to simplify, we call these points extrema). To find them, for an arbitrary set of points, it is necessary to browse all the points but not in the particular case of a binary shape. In this specific situation, the extrema necessarily belong to the inner boundary of the shape. It is a waste of time to browse all the pixels as it is possible to discard most of them when we search for these extrema. Based on this analysis, we propose a new method to compute the convex hull dedicated to binary shapes. This method browses as few pixels as possible to select a small subset of boundary pixels. Then it deduces the convex hull only from this subset. As the size of the subset is very small, the convex hull is computed in real time. We compare it with the commonly used methods and common functions from libraries to prove that our approach is faster. This comparison shows that, for a very small shape, the difference is acceptable, but when the area of the shape grows, this difference becomes significant. This leads us to conclude that substituting current functions to compute convex hull of binary shapes with our algorithm in frequently used libraries would lead to a great improvement.
Citations : 0
Bibtex
[+] "The Dahu Graph-Cut for Interactive Segmentation on 2D/3D Images", Ôn Vũ Ngọc, Edwin Carlinet, Jonathan Fabrizio, Thierry GéraudIn this article, we present an algorithm to compute the convex hull of a binary shape. Efficient algorithms to compute the convex hull of a set of points had been proposed long time ago. For a binary shape, the common practice is to rely on one of them: to compute the convex hull of binary shape, all pixels of the shape are first listed, and then the convex hull is computed on this list of points. The computed convex hull is finally rasterized to provide the final result (as, for example, in the famous scikit-image library). To compute the convex hull of an arbitrary set of points, the points of the list that lie on the outline of the convex hull must be selected (to simplify, we call these points extrema). To find them, for an arbitrary set of points, it is necessary to browse all the points but not in the particular case of a binary shape. In this specific situation, the extrema necessarily belong to the inner boundary of the shape. It is a waste of time to browse all the pixels as it is possible to discard most of them when we search for these extrema. Based on this analysis, we propose a new method to compute the convex hull dedicated to binary shapes. This method browses as few pixels as possible to select a small subset of boundary pixels. Then it deduces the convex hull only from this subset. As the size of the subset is very small, the convex hull is computed in real time. We compare it with the commonly used methods and common functions from libraries to prove that our approach is faster. This comparison shows that, for a very small shape, the difference is acceptable, but when the area of the shape grows, this difference becomes significant. This leads us to conclude that substituting current functions to compute convex hull of binary shapes with our algorithm in frequently used libraries would lead to a great improvement.
Citations : 0
Bibtex
@article{fabrizio-jrtip23, author = {Fabrizio, Jonathan}, title = {How to Compute the Convex Hull of a Binary Shape? A Real-Time Algorithm to Compute the Convex Hull of a Binary Shape}, year = {2023}, issue_date = {Dec 2023}, volume = {20}, number = {6}, issn = {1861-8200}, doi = {10.1007/s11554-023-01359-8}, journal = {J. Real-Time Image Processing}, month = {sep}, numpages = {12}, }
(PR23), Pattern Recognition, Volume 136, Number 109207, 2023
Abstract :
Interactive image segmentation is an important application in computer vision for selecting objects of interest in images. Several interactive segmentation methods are based on distance transform algorithms. However, the most known distance transform, geodesic distance, is sensitive to noise in the image and to seed placement. Recently, the Dahu pseudo-distance, a continuous version of the minimum barrier distance (MBD), is proved to be more powerful than the geodesic distance in noisy and blurred images. This paper presents a method for combining the Dahu pseudo-distance with edge information in a graph-cut optimization framework and leveraging each's complementary strengths. Our method works efficiently on both 2D/3D images and videos. Results show that our method achieves better performance than other distance-based and graph-cut methods, thereby reducing the user's efforts.
Citations : 0
Bibtex
[+] "A New Minimum Barrier Distance for Multivariate Images with Applications to Salient Object Detection, Shortest Path Finding, and Segmentation", Minh Ôn Vũ Ngọc, Nicolas Boutry, Jonathan Fabrizio, Thierry GéraudInteractive image segmentation is an important application in computer vision for selecting objects of interest in images. Several interactive segmentation methods are based on distance transform algorithms. However, the most known distance transform, geodesic distance, is sensitive to noise in the image and to seed placement. Recently, the Dahu pseudo-distance, a continuous version of the minimum barrier distance (MBD), is proved to be more powerful than the geodesic distance in noisy and blurred images. This paper presents a method for combining the Dahu pseudo-distance with edge information in a graph-cut optimization framework and leveraging each's complementary strengths. Our method works efficiently on both 2D/3D images and videos. Results show that our method achieves better performance than other distance-based and graph-cut methods, thereby reducing the user's efforts.
Citations : 0
Bibtex
@Article{ movn-pr22, author = {Minh \^On V\~{u} Ng\d{o}c and Edwin Carlinet and Jonathan Fabrizio and Thierry G\'eraud}, title = {The {D}ahu Graph-Cut for Interactive Segmentation on {2D/3D} Images}, journal = {Pattern Recognition}, year = {2023}, volume = {136}, number = {109207}, month = apr, abstract = {Interactive image segmentation is an important application in computer vision for selecting objects of interest in images. Several interactive segmentation methods are based on distance transform algorithms. However, the most known distance transform, geodesic distance, is sensitive to noise in the image and to seed placement. Recently, the Dahu pseudo-distance, a continuous version of the minimum barrier distance (MBD), is proved to be more powerful than the geodesic distance in noisy and blurred images. This paper presents a method for combining the Dahu pseudo-distance with edge information in a graph-cut optimization framework and leveraging each's complementary strengths. Our method works efficiently on both 2D/3D images and videos. Results show that our method achieves better performance than other distance-based and graph-cut methods, thereby reducing the user's efforts.}, doi = {10.1016/j.patcog.2022.109207} url = {http://jo.fabrizio.free.fr/research/publi/pdf/movn_pr22.pdf} }
Computer Vision and Image Understanding (CVIU20), Volume 197-198, 2020
Abstract :
Distance transforms and the saliency maps they induce are widely used in image processing, computer vision, and pattern recognition. One of the most commonly used distance transform is the geodesic one. Unfortunatelythis distance does not always achieve satisfying results on noisy or blurred images. Recently, a new (pseudo-)distance, called the minimum barrier distance (MBD), more robust to pixel variations, has been introduced. Some years after, Géraud et al. have proposed a good and fast-to compute approximation of this distance: the Dahu pseudo-distance. Since this distance was initially developped for grayscale images, we propose here an extension of this transform to multivariate images; we call it vectorial Dahu pseudo-distance. An efficient way to compute it is provided in this paper. Besides, we provide benchmarks demonstrating how much the vectorial Dahu pseudo-distance is more robust and competitive compared to other MB-based distances, which shows how much this distance is promising for salient object detectionshortest path finding, and object segmentation.
Citations : 0
Bibtex
[+] "TextCatcher: a Method to Detect Curved and Challenging Text in Natural Scenes", J. Fabrizio, M. Robert-Seidowsky, S. Dubuisson, A. Calarasanu, A. and R. BoisselDistance transforms and the saliency maps they induce are widely used in image processing, computer vision, and pattern recognition. One of the most commonly used distance transform is the geodesic one. Unfortunatelythis distance does not always achieve satisfying results on noisy or blurred images. Recently, a new (pseudo-)distance, called the minimum barrier distance (MBD), more robust to pixel variations, has been introduced. Some years after, Géraud et al. have proposed a good and fast-to compute approximation of this distance: the Dahu pseudo-distance. Since this distance was initially developped for grayscale images, we propose here an extension of this transform to multivariate images; we call it vectorial Dahu pseudo-distance. An efficient way to compute it is provided in this paper. Besides, we provide benchmarks demonstrating how much the vectorial Dahu pseudo-distance is more robust and competitive compared to other MB-based distances, which shows how much this distance is promising for salient object detectionshortest path finding, and object segmentation.
Citations : 0
Bibtex
@article{movn-cviu20, author = {Minh {\^On V\~{u} Ng\d{o}c} and Nicolas Boutry and Jonathan Fabrizio and Thierry G\'eraud}, title = {A New Minimum Barrier Distance for Multivariate Images with Applications to Salient Object Detection, Shortest Path Finding, and Segmentation}, journal = {Computer Vision and Image Understanding}, volume = {197--198}, month = aug, year = {2020}, }
International Journal an Document Analysis and Recognition (IJDAR16), Volume 19, 2016, Pages 99–117
Abstract :
In this paper, we propose a text detection algorithm which is hybrid and multi-scale. First, it relies on a connected component-based approach: After the segmentation of the image, a classification step using a new wavelet descriptor spots the letters. A new graph modeling and its traversal procedure allow to form candidate text areas. Second, a texture-based approach discards the false positives. Finally, the detected text areas are precisely cut out and a new binarization step is introduced. The main advantage of our method is that few assumptions are put forward. Thus, “challenging texts” like multi-sized, multi-colored, multi-oriented or curved text can be localized. The efficiency of TextCatcher has been validated on three different datasets: Two come from the ICDAR competition, and the third one contains photographs we have taken with various daily life texts. We present both qualitative and quantitative results.
Citations : 4
Bibtex
[+] "What is a good evaluation protocol for text localization systems? Concerns, arguments, comparisons and solutions", A. Calarasanu, J. Fabrizio and S. DubuissonIn this paper, we propose a text detection algorithm which is hybrid and multi-scale. First, it relies on a connected component-based approach: After the segmentation of the image, a classification step using a new wavelet descriptor spots the letters. A new graph modeling and its traversal procedure allow to form candidate text areas. Second, a texture-based approach discards the false positives. Finally, the detected text areas are precisely cut out and a new binarization step is introduced. The main advantage of our method is that few assumptions are put forward. Thus, “challenging texts” like multi-sized, multi-colored, multi-oriented or curved text can be localized. The efficiency of TextCatcher has been validated on three different datasets: Two come from the ICDAR competition, and the third one contains photographs we have taken with various daily life texts. We present both qualitative and quantitative results.
Citations : 4
- "A tutorial on well-composedness", N Boutry, T Géraud, L Najman (JMIV 2018)
- "Scene text detection based on enhanced multi-channels MSER and a fast text grouping process", J Dai, Z Wang, X Zhao, S Shao (ICCCBDA 2018)
- "Morphology-based hierarchical representation with application to text segmentation in natural images", LD Huỳnh, Y Xu, T Géraud (ICPR 2016)
- "From text detection to text segmentation: a unified evaluation scheme", S Calarasanu, J Fabrizio, S Dubuisson (IWRR-ECCV 2016)
Bibtex
@article{fabrizio-ijdar16, author = {Jonathan Fabrizio and Myriam Robert-Seidowsky and S{\'e}verine Dubuisson and Stefania Calarasanu and Rapha{\"e}l Boissel}, title = {TextCatcher: a method to detect curved and challenging text in natural scenes}, booktitle = {International Journal on Document Analysis and Recognition (IJDAR)}, volume = {19}, pages = {99--117}, year = {2016}, }
Image and Vision Computing (IVC16), Volume 46, 2016
Abstract :
A trustworthy protocol is essential to evaluate a text detection algorithm in order to, first measure its efficiency and adjust its parameters and, second to compare its performances with those of other algorithms. However, current protocols do not give precise enough evaluations because they use coarse evaluation metrics, and deal with inconsistent matchings between the output of detection algorithms and the ground truth, both often limited to rectangular shapes. In this paper, we propose a new evaluation protocol, named EvaLTex, that solves some of the current problems associated with classical metrics and matching strategies. Our system deals with different kinds of annotations and detection shapes. It also considers different kinds of granularity between detections and ground truth objects and hence provides more realistic and accurate evaluation measures. We use this protocol to evaluate text detection algorithms and highlight some key examples that show that the provided scores are more relevant than those of currently used evaluation protocols. We propose a two-level annotation evaluation protocol for text detection algorithms.Algorithms with different granularity outputs are equitably compared.All matching strategies between the ground truth and the detections are handled.Quantity and quality scores are given to describe a detector's behavior.The protocol can manage any irregular text representation.
Citations : 4
Bibtex
[+] "Text detection in street level images", J. Fabrizio, B. Marcotegui and M. CordA trustworthy protocol is essential to evaluate a text detection algorithm in order to, first measure its efficiency and adjust its parameters and, second to compare its performances with those of other algorithms. However, current protocols do not give precise enough evaluations because they use coarse evaluation metrics, and deal with inconsistent matchings between the output of detection algorithms and the ground truth, both often limited to rectangular shapes. In this paper, we propose a new evaluation protocol, named EvaLTex, that solves some of the current problems associated with classical metrics and matching strategies. Our system deals with different kinds of annotations and detection shapes. It also considers different kinds of granularity between detections and ground truth objects and hence provides more realistic and accurate evaluation measures. We use this protocol to evaluate text detection algorithms and highlight some key examples that show that the provided scores are more relevant than those of currently used evaluation protocols. We propose a two-level annotation evaluation protocol for text detection algorithms.Algorithms with different granularity outputs are equitably compared.All matching strategies between the ground truth and the detections are handled.Quantity and quality scores are given to describe a detector's behavior.The protocol can manage any irregular text representation.
Citations : 4
- "Morphology-based hierarchical representation with application to text segmentation in natural images", LD Huỳnh, Y Xu, T Géraud (ICPR16)
- "Integrating scene text and visual appearance for fine-grained image classification", X Bai, M Yang, P Lyu, Y Xu, J Luo
- "From text detection to text segmentation: a unified evaluation scheme", S Calarasanu, J Fabrizio, S Dubuisson ()
- "Quality-Driven video analysis for the improvement of foreground segmentation", D Ortego Hernández (PhD 2018)
Bibtex
@article{fabrizio-ivc16, author = {Stefania Calarasanu and Jonathan Fabrizio and Severine Dubuisson}, title = {What is a Good Evaluation Protocol for Text Localization Systems? Concerns, Arguments, Comparisons and Solutions}, booktitle = {Image and Vision Computing}, volume = {46}, pages = {1-17}, year = {2016}, issn = {0262-8856}, year = {2016}, doi = {https://doi.org/10.1016/j.imavis.2015.12.001}, url = {http://jo.fabrizio.free.fr/research/publi/pdf/calarasanu_ivc16}, }
Pattern Analysis and Applications (PAA13), Volume 16, Issue 4, nov 2013, Pages 519–533
Abstract :
Text detection system for natural images is a very challenging task in Computer Vision. Image acquisition introduces distortion in terms of perspective, blurring, illumination, and characters which may have very different shape, size, and color. We introduce in this article a full text detection scheme. Our architecture is based on a new process to combine a hypothesis generation step to get potential boxes of text and a hypothesis validation step to filter false detections. The hypothesis generation process relies on a new efficient segmentation method based on a morphological operator. Regions are then filtered and classified using shape descriptors based on Fourier, Pseudo Zernike moments and an original polar descriptor, which is invariant to rotation. Classification process relies on three SVM classifiers combined in a late fusion scheme. Detected characters are finally grouped to generate our text box hypotheses. Validation step is based on a global SVM classification of the box content using dedicated descriptors adapted from the HOG approach. Results on the well-known ICDAR database are reported showing that our method is competitive. Evaluation protocol and metrics are deeply discussed and results on a very challenging street-level database are also proposed.
Citations : 0
Bibtex
[+] "Motion compensation based on Tangent Distance prediction for video compression", J. Fabrizio, S. Dubuisson and D. BéréziatText detection system for natural images is a very challenging task in Computer Vision. Image acquisition introduces distortion in terms of perspective, blurring, illumination, and characters which may have very different shape, size, and color. We introduce in this article a full text detection scheme. Our architecture is based on a new process to combine a hypothesis generation step to get potential boxes of text and a hypothesis validation step to filter false detections. The hypothesis generation process relies on a new efficient segmentation method based on a morphological operator. Regions are then filtered and classified using shape descriptors based on Fourier, Pseudo Zernike moments and an original polar descriptor, which is invariant to rotation. Classification process relies on three SVM classifiers combined in a late fusion scheme. Detected characters are finally grouped to generate our text box hypotheses. Validation step is based on a global SVM classification of the box content using dedicated descriptors adapted from the HOG approach. Results on the well-known ICDAR database are reported showing that our method is competitive. Evaluation protocol and metrics are deeply discussed and results on a very challenging street-level database are also proposed.
Citations : 0
Bibtex
@article{fabrizio-paa13, title = "Text detection in street level images", author = "Jonathan Fabrizio and Beatriz Marcotegui and Matthieu Cord", journal = "Pattern Analysis and Applications", volume = "16", number = "4", pages = "519--533", year = "2013", }
Signal Processing: Image Communication (SPIC12), Volume 27, Issue 2, February 2012, Pages 153–171
Abstract :
We present a new algorithm for motion compensation that uses a motion estimation method based on tangent distance. The method is compared with a Block-Matching based approach in various common situations. Whereas Block-Matching algorithms usually only predict positions of blocks over time, our method also predicts the evolution of pixels into these blocks. The prediction error is then drastically decreased. The method is implemented into the Theora codec proving that this algorithm improves the video codec performances.
Citations : 1
Bibtex
[+] "Optimal recursive clustering of likelihood functions for multiple object tracking", S. Dubuisson and J. FabrizioWe present a new algorithm for motion compensation that uses a motion estimation method based on tangent distance. The method is compared with a Block-Matching based approach in various common situations. Whereas Block-Matching algorithms usually only predict positions of blocks over time, our method also predicts the evolution of pixels into these blocks. The prediction error is then drastically decreased. The method is implemented into the Theora codec proving that this algorithm improves the video codec performances.
Citations : 1
- "CODAGE DES DONNÉES VISUELLES: EFFICACITÉ, ROBUSTESSE, TRANSMISSION", Marco Cagnazzo 2013 (HDR)
Bibtex
@article{fabrizio-spic12, title = "Motion compensation based on tangent distance prediction for video compression", author = "Jonathan Fabrizio and S{\'e}verine Dubuisson and Dominique B{\'e}r{\'e}ziat", journal = "Signal Processing: Image Communication", volume = "27", number = "2", pages = "153--171", year = "2012", issn = "0923-5965", doi = "10.1016/j.image.2011.12.001", }
Pattern Recognition Letters (PRL09), Volume: 30, Issue: 6, April 2009.
Abstract :
In this paper, we propose a method to track multiple deformable objects in sequences (with a static camera) in and beyond the visible spectrum by combining Gabor filtering and clustering. The idea is to sample moving areas between two frames by randomly positioning samples over high magnitude area of a motion likelihood function. These points are then clustered to obtain one class for each moving object. The novelty in our method is in using cluster information from the previous frame to classify new samples in the current frame: we call that a recursive clustering. This makes our method robust to occlusions, objects entering and leaving the field of view, objects stopping and starting, and moving objects getting really close to each other.
Citations : 2
Bibtex
[+] "An Analytical Solution to the Perspective-N-Point Problem for Common Planar Camera and for Catadioptric Sensor", J. Fabrizio, J. DevarsIn this paper, we propose a method to track multiple deformable objects in sequences (with a static camera) in and beyond the visible spectrum by combining Gabor filtering and clustering. The idea is to sample moving areas between two frames by randomly positioning samples over high magnitude area of a motion likelihood function. These points are then clustered to obtain one class for each moving object. The novelty in our method is in using cluster information from the previous frame to classify new samples in the current frame: we call that a recursive clustering. This makes our method robust to occlusions, objects entering and leaving the field of view, objects stopping and starting, and moving objects getting really close to each other.
Citations : 2
- "A multi-cue spatio-temporal framework for automatic frontal face clustering in video sequences", Simeon Schwab, Thierry Chateau, Christophe Blanc and Laurent Trassoudaine, EURASIP Journal on Image and Video Processing (2013)
- "Multi-Person Tracking Using a Discriminative Color Appearance Model", Siddharth Chunduri - Master Thesis (2012)
Bibtex
@article{dubuisson-prl09, author = {Dubuisson, S\'{e}verine and Fabrizio, Jonathan}, title = {Optimal recursive clustering of likelihood functions for multiple object tracking}, journal = {Pattern Recogn. Lett.}, volume = {30}, issue = {6}, month = {April}, year = {2009}, issn = {0167-8655}, pages = {606--614}, numpages = {9}, doi = {10.1016/j.patrec.2009.01.001}, publisher = {Elsevier Science Inc.}, address = {New York, NY, USA}, }
International Journal of Image and Graphics (IJIG08), Volume: 8, Issue: 1 (January 2008)
Abstract :
The Perspective-N-Point problem (PNP) is a notable problem in computer vision. It consists of given N points known in an object coordinate space and their projection onto the image, estimating the distance between the video camera and the set of points. By the use of an unusual formulation, we propose a method to get a strictly analytical solution based on the resolution of linear systems. This solution can be computed instantly and is well adapted to real time computer vision applications. Our approach is general enough to work with a nonlinear sensor like a catadioptric panoramic sensor. To improve the localization accuracy, we also provide a technique to correct geometrical distortion. This algorithm also corrects little errors on intrinsic and extrinsic parameters. Well implemented, this correction can be performed in real time.
Citations : 6
Bibtex
[+] "Estimation numérique des courbes épipolaires pour les capteurs omnidirectionnels", J. Fabrizio, J. DevarsThe Perspective-N-Point problem (PNP) is a notable problem in computer vision. It consists of given N points known in an object coordinate space and their projection onto the image, estimating the distance between the video camera and the set of points. By the use of an unusual formulation, we propose a method to get a strictly analytical solution based on the resolution of linear systems. This solution can be computed instantly and is well adapted to real time computer vision applications. Our approach is general enough to work with a nonlinear sensor like a catadioptric panoramic sensor. To improve the localization accuracy, we also provide a technique to correct geometrical distortion. This algorithm also corrects little errors on intrinsic and extrinsic parameters. Well implemented, this correction can be performed in real time.
Citations : 6
- "Camera Models and Fundamental Concepts Used in Geometric Computer Vision", Peter Sturm, Srikumar Ramalingam, Jean-Philippe Tardif, Simone Gasparini and João Barreto (FTCGV 2011)
- ? "The Visual SLAM System for a Hexapod Robot", Adam Schmidt and Andrzej Kasiński (ICCVG 2010)*
- "Infrared Based Camera Registration for In-Door/Out-Door Augmented Reality", Vincent Guitteny, Oussama Moslah, Serge Couvet (AFRV 2008)
- "Infrared Camera Registration for Urban Augmented Reality", Vincent GUITTENY, François-Pierre ROBERT and Oussama MOSLAH (REGARD 2008)
- "Multiple-view vision-based robot calibration", Kristoer Rath Petersen Michael Bing (Master's Thesis 2012)
- "Development of a Calibration Module For Stereoscopic Configurations", Satyam Gupta (2013)
Bibtex
@article{fabrizio-ijig08, author = {Fabrizio, J. and Devars, J.}, title = {An Analytical Solution to the Perspective-N-Point Problem for Common Planar Camera and for Catadioptric Sensor}, booktitle = {International Journal of Image and Graphics}, volume = {8}, number = {1} date = {January}, pages = {135-155}, year = {2008}, }
Revue Traitement du Signal (TS05), Special Issue on Omnidirectional Vision, Volume 22, Number 5, pages 527-535, December 2005
Abstract :
The epipolar geometry of couples of omnidirectional sensors is often difficult to express analytically. We propose an algorithm to estimate numerically epipolar curves from omnidirectional pairs of stereovision. This algorithm is not limited to this type of sensors and works, for example, with a combination of a panoramic sensor and a traditional camera. Although the load of calculation necessary for this algorithm is heavy, it works with every kind of sensor (provided that the stereovision pair is completely calibrated) especially with sensor that do not respect the single viewpoint constraint.
Citations : 2
Bibtex
The epipolar geometry of couples of omnidirectional sensors is often difficult to express analytically. We propose an algorithm to estimate numerically epipolar curves from omnidirectional pairs of stereovision. This algorithm is not limited to this type of sensors and works, for example, with a combination of a panoramic sensor and a traditional camera. Although the load of calculation necessary for this algorithm is heavy, it works with every kind of sensor (provided that the stereovision pair is completely calibrated) especially with sensor that do not respect the single viewpoint constraint.
Citations : 2
- "A direct model for non central catadioptric cameras with quadric shaped mirrors", Bertrand Vandeportaele, Pierre Gurdjos (ORASIS 2007)
- "Contributions à la vision omnidirectionnelle: Étude, Conception et Étalonnage de capteurs pour l'acquisition d'images et la modélisation 3D", Bertrand VANDEPORTAELE (PhD Thesis 2006)
Bibtex
@article{fabrizio-ts05, author = {Fabrizio, J. and Devars, J.}, title = {Estimation numerique des courbes epipolaires pour les capteurs omnidirectionnels}, booktitle = {Revue Traitement du Signal, Special Issue on Omnidirectionnal Vision}, volume = {22}, number = {5} date = {December}, pages = {527-535}, year = {2005}, url = {http://jo.fabrizio.free.fr/recherche/fabdev_ts05.pdf} }
Conference papers
[+] Saliency-based detection of identity documents captured by smartphones, Minh Ôn Vũ Ngọc, Jonathan Fabrizio, Thierry Géraud13th IAPR International Workshop on Document Analysis Systems (DAS18)
Abstract :
Smartphones have became an easy and convenient mean to acquire documents. In this paper, we focus on the automatic segmentation of identity documents in smartphone photos or videos using visual saliency (VS). VS-based approaches, which pertain to computer vision, have not be considered yet for this particular task. Here we compare different VS meth- ods, and we propose a new VS scheme, based on a recent distance belonging to the scope of mathematical morphology. We show that our resulting saliency maps are competitive with state-of-the-art visual saliency methods, and that such approaches are very promising for use in identity document detection and segmentation, even without taking into account any prior knowledge about document contents. In particular they can perform in real-time on smartphones.
Citations : 0
Bibtex
[+] A first step toward a fair comparison of evaluation protocols for text detection algorithms, Aliona Dangla, Elodie Puybareau, Guillaume Tochon, Jonathan FabrizioSmartphones have became an easy and convenient mean to acquire documents. In this paper, we focus on the automatic segmentation of identity documents in smartphone photos or videos using visual saliency (VS). VS-based approaches, which pertain to computer vision, have not be considered yet for this particular task. Here we compare different VS meth- ods, and we propose a new VS scheme, based on a recent distance belonging to the scope of mathematical morphology. We show that our resulting saliency maps are competitive with state-of-the-art visual saliency methods, and that such approaches are very promising for use in identity document detection and segmentation, even without taking into account any prior knowledge about document contents. In particular they can perform in real-time on smartphones.
Citations : 0
Bibtex
@article{movn-das18, author = {M. Ô. V. Ngoc and J. Fabrizio and T. Géraud}, title = {Saliency-Based Detection of Identy Documents Captured by Smartphones}, booktitle = {2018 13th IAPR International Workshop on Document Analysis Systems (DAS)}, pages = {387-392}, year = {2018}, month = {April}, }
13th IAPR International Workshop on Document Analysis Systems (DAS18)
Abstract :
Text detection is an important topic in pattern recognition, but evaluating the reliability of such detection algorithms is challenging. While many evaluation protocols have been developed for that purpose, they often show dissimilar behaviors when applied in the same context. As a consequence, their usage may lead to misinterpretations, potentially yielding erroneous comparisons between detection algorithms or their incorrect parameters tuning. This paper is a first attempt to derive a methodology to perform the comparison of evaluation protocols. We then apply it on five state-of-the-art protocols, and exhibit that there indeed exist inconsistencies among their evaluation criteria. Our aim here is not to rank the investigated evaluation protocols, but rather raising awareness in the community that we should carefully reconsider them in order to converge to their optimal usage.
Citations : 0
Bibtex
[+] "Towards the rectification of highly distorted texts", Stefania Calarasanu and S\'everine Dubuisson and Jonathan FabrizioText detection is an important topic in pattern recognition, but evaluating the reliability of such detection algorithms is challenging. While many evaluation protocols have been developed for that purpose, they often show dissimilar behaviors when applied in the same context. As a consequence, their usage may lead to misinterpretations, potentially yielding erroneous comparisons between detection algorithms or their incorrect parameters tuning. This paper is a first attempt to derive a methodology to perform the comparison of evaluation protocols. We then apply it on five state-of-the-art protocols, and exhibit that there indeed exist inconsistencies among their evaluation criteria. Our aim here is not to rank the investigated evaluation protocols, but rather raising awareness in the community that we should carefully reconsider them in order to converge to their optimal usage.
Citations : 0
Bibtex
@article{fabrizio-xxx, author = {}, title = {}, booktitle = {}, volume = {}, number = {} date = {}, pages = {}, year = {}, }
International Conference on Computer Vision Theory and Applications (VISAPP16)
Abstract :
A frequent challenge for many Text Understanding Systems is to tackle the variety of text characteristics in born-digital and natural scene images to which current OCRs are not well adapted. For example, texts in perspective are frequently present in real-word images, but despite the ability of some detectors to accurately localize such text objects, the recognition stage fails most of the time. Indeed, most OCRs are not designed to handle text strings in perspective but rather expect horizontal texts in a parallel-frontal plane to provide a correct transcription. In this paper, we propose a rectification procedure that can correct highly distorted texts, subject to rotation, shearing and perspective deformations. The method is based on an accurate estimation of the quadrangle bounding the deformed text in order to compute a homography to transform this quadrangle (and its content) into a horizontal rectangle. The rectification is validated on the dataset proposed during the ICDAR 2015 Competition on Scene Text Rectification.
Citations : 0
Bibtex
[+] Using histogram representation and Earth Mover's Distance as an evaluation tool for text detection, Stefania Calarasanu, Jonathan Fabrizio, Séverine DubuissonA frequent challenge for many Text Understanding Systems is to tackle the variety of text characteristics in born-digital and natural scene images to which current OCRs are not well adapted. For example, texts in perspective are frequently present in real-word images, but despite the ability of some detectors to accurately localize such text objects, the recognition stage fails most of the time. Indeed, most OCRs are not designed to handle text strings in perspective but rather expect horizontal texts in a parallel-frontal plane to provide a correct transcription. In this paper, we propose a rectification procedure that can correct highly distorted texts, subject to rotation, shearing and perspective deformations. The method is based on an accurate estimation of the quadrangle bounding the deformed text in order to compute a homography to transform this quadrangle (and its content) into a horizontal rectangle. The rectification is validated on the dataset proposed during the ICDAR 2015 Competition on Scene Text Rectification.
Citations : 0
Bibtex
@InProceedings{calarasanu-visapp16, author = {Stefania Calarasanu and S\'everine Dubuisson and Jonathan Fabrizio}, title = {Towards the rectification of highly distorted texts}, booktitle = {Proceedings of the 11th International Conference on Computer Vision Theory and Applications (VISAPP)}, address = {Rome, Italie}, month = feb, year = 2016 }
Proceedings of the 13th IAPR International Conference on Document Analysis and Recognition (ICDAR15)
Abstract :
In the context of text detection evaluation, it is essential to use protocols that are capable of describing both the quality and the quantity aspects of detection results. In this paper we propose a novel visual representation and evaluation tool that captures the whole nature of a detector by using histograms. First, two histograms (coverage and accuracy) are generated to visualize the different characteristics of a detector. Secondly, we compare these two histograms to a so called optimal one to compute representative and comparable scores. To do so, we introduce the usage of the Earth Mover's Distance as a reliable evaluation tool to estimate recall and precision scores. Results obtained on the ICDAR 2013 dataset show that this method intuitively characterizes the accuracy of a text detector and gives at a glance various useful characteristics of the analyzed algorithm.
Citations : 4
Bibtex
[+] TEXT TRAIL: A Robust Text Tracking Algorithm In Wild Environments, Myriam Robert-Seidowsky, Jonathan Fabrizio and Séverine DubuissonIn the context of text detection evaluation, it is essential to use protocols that are capable of describing both the quality and the quantity aspects of detection results. In this paper we propose a novel visual representation and evaluation tool that captures the whole nature of a detector by using histograms. First, two histograms (coverage and accuracy) are generated to visualize the different characteristics of a detector. Secondly, we compare these two histograms to a so called optimal one to compute representative and comparable scores. To do so, we introduce the usage of the Earth Mover's Distance as a reliable evaluation tool to estimate recall and precision scores. Results obtained on the ICDAR 2013 dataset show that this method intuitively characterizes the accuracy of a text detector and gives at a glance various useful characteristics of the analyzed algorithm.
Citations : 4
- "Morphology-based hierarchical representation with application to text segmentation in natural images"
- "From text detection to text segmentation: a unified evaluation scheme"
- "Histogram optimization with CUDA"
- "Towards the Rectification of Highly Distorted Texts"
Bibtex
@inproceedings{calarasanu-icdar15, author = {S. Calarasanu and J. Fabrizio and S. Dubuisson}, title = {Using histogram representation and Earth Mover's Distance as an evaluation tool for text detection}, booktitle = {2015 13th International Conference on Document Analysis and Recognition (ICDAR)}, pages = {221-225}, year = {2015}, month = {Aug} }
VISAPP2015
Abstract :
Citations : 0
Bibtex
[+] A self-adaptive likelihood function for tracking with particle filter, Séverine Dubuisson, Myriam Robert-Seidowsky and Jonathan FabrizioCitations : 0
Bibtex
@article{Robert-Seidowsky-14, author = {}, title = {}, booktitle = {}, volume = {}, number = {} date = {}, pages = {}, year = {}, }
VISAPP2015
Abstract :
Citations : 0
Bibtex
[+] A Precise Skew Estimation Algorithm for Document Images Using KNN Clustering and Fourier Transform, J. FabrizioCitations : 0
Bibtex
@article{dubuisson-xxx, author = {}, title = {}, booktitle = {}, volume = {}, number = {} date = {}, pages = {}, year = {}, }
IEEE International Conference on Image Processing (ICIP14). Paris, France. Oct 27-30 2014.
Abstract :
In this article, we propose a simple and precise skew estimation algorithm for binarized document images. The estimation is performed in the frequency domain. To get a precise result, the Fourier transform is not applied to the document itself but the document is preprocessed: all regions of the document are clustered using a KNN and contours of grouped regions are smoothed using the convex hull to form more regular shapes, with better orientation. No assumption has been made concerning the nature or the content of the document. This method has been shown to be very accurate and was ranked first at the DISEC'13 contest, during the ICDAR competitions.
Citations : 0
Bibtex
[+] Snoopertext: A multiresolution system for text detection in complex visual scenes, R. Minetto, N. Thome, M. Cord, J. Fabrizio and B. Marcotegui.In this article, we propose a simple and precise skew estimation algorithm for binarized document images. The estimation is performed in the frequency domain. To get a precise result, the Fourier transform is not applied to the document itself but the document is preprocessed: all regions of the document are clustered using a KNN and contours of grouped regions are smoothed using the convex hull to form more regular shapes, with better orientation. No assumption has been made concerning the nature or the content of the document. This method has been shown to be very accurate and was ranked first at the DISEC'13 contest, during the ICDAR competitions.
Citations : 0
Bibtex
@article{fabrizio-14, author = {Jonathan Fabrizio}, title = {A Precise Skew Estimation Algorithm for Document Images Using KNN Clustering and Fourier Transform.}, booktitle = {IEEE International Conference on Image Processing}, volume = {}, number = {} date = {}, pages = {}, year = {2014}, }
IEEE International Conference on Image Processing (ICIP 10). September 26-29, 2010. Hong Kong.
Abstract :
Text detection in natural images remains a very challenging task. For instance, in an urban context, the detection is very difficult due to large variations in terms of shape, size, color, orientation, and the image may be blurred or have irregular illumination, etc. In this paper, we describe a robust and accurate multiresolution approach to detect and classify text regions in such scenarios. Based on generation/validation paradigm, we first segment images to detect character regions with a multiresolution algorithm able to manage large character size variations. The segmented regions are then filtered out using shapebased classification, and neighboring characters are merged to generate text hypotheses. A validation step computes a region signature based on texture analysis to reject false positives. We evaluate our algorithm in two challenging databases, achieving very good result
Citations : 3
Bibtex
[+] Text Segmentation in Natural Scenes Using Toggle-Mapping, J. Fabrizio, M. Beatriz and M. CordText detection in natural images remains a very challenging task. For instance, in an urban context, the detection is very difficult due to large variations in terms of shape, size, color, orientation, and the image may be blurred or have irregular illumination, etc. In this paper, we describe a robust and accurate multiresolution approach to detect and classify text regions in such scenarios. Based on generation/validation paradigm, we first segment images to detect character regions with a multiresolution algorithm able to manage large character size variations. The segmented regions are then filtered out using shapebased classification, and neighboring characters are merged to generate text hypotheses. A validation step computes a region signature based on texture analysis to reject false positives. We evaluate our algorithm in two challenging databases, achieving very good result
Citations : 3
- "Robust text detection in natural images with edge-enhanced maximally stable extremal regions", Huizhong Chen, Sam S. Tsai, Georg Schroth, David M. Chen, Radek Grzeszczuk and Bernd Girod (ICIP 2011)
- "SNOOPERTRACK: Text detection and tracking for outdoor videos", R. Minetto, N. Thome, M. Cord, N.J Leite, J. Stolfi (ICIP 2011)
- "Text Detection and Recognition in Urban Scenes", R. Minetto, N. Thome, M. Cord, J. Stolfi, F. Pr´ecioso, J. Guyomard, N.J Leite (CVRS workshop - ICCV 2011)
Bibtex
@inproceedings{-, author = {Rodrigo Minetto and Nicolas Thome and Matthieu Cord and Jonathan Fabrizio and Beatriz Marcotegui}, title = {Snoopertext: A multiresolution system for text detection in complex visual scenes}, booktitle = {ICIP}, date = {September 26-29}, pages = {3861-3864}, address = {Hong Kong, China}, year = {2010}, }
IEEE International Conference on Image Processing (ICIP09). November 7-10, 2009. Cairo, Egypt.
Abstract :
We offer, in this paper, a new method to segment text in natural scenes. This method is based on the use of a morphological operator: the Toggle Mapping. The efficiency of the method is illustrated and the method is compared, according to various criteria, with common methods issued from the state of the art. This comparison shows that our method gives better results and is faster than the state of the art methods. Our method reduces also the number of segmented regions. This can lead to time saving in a complete scheme (executing time of multiple processing steps usually depends on the number of regions) and proves that our algorithm is more relevant.
Citations : 10
Bibtex
[+] Text Extraction from Street Level Images, J. Fabrizio, M. Cord and B. MarcoteguiWe offer, in this paper, a new method to segment text in natural scenes. This method is based on the use of a morphological operator: the Toggle Mapping. The efficiency of the method is illustrated and the method is compared, according to various criteria, with common methods issued from the state of the art. This comparison shows that our method gives better results and is faster than the state of the art methods. Our method reduces also the number of segmented regions. This can lead to time saving in a complete scheme (executing time of multiple processing steps usually depends on the number of regions) and proves that our algorithm is more relevant.
Citations : 10
- "Super-resolved binarization of text based on the FAIR algorithm", Thibault Lelore, Frédéric Bouchara (ICDAR 2011)
- "Local Thresholding Algorithm Based on Variable Window Size Statistics", Costin-Anton Boiangiu, Alexandra Olteanu, Alexandru Stefanescu, Daniel Rosner, Nicolae Tapus, Mugurel Andreica (CSCS 2011)
- "DIBCO 2009: document image binarization contest", B. Gatos, K. Ntirogiannis, I. Pratikakis (IJDAR 2011)
- "Adaptative Binarization Method With Variable Window Size - A Parameter Free Approach", Alexandra Olteanu, Alexandru Victor Stefanescu, Costin-Anton Boiangiu (GHC 2010)
- "H-DIBCO 2010: Handwritten Document Image Binarization Competition", Ioannis Pratikakis, Basilis Gatos and Konstantinos Ntirogiannis (H-DIBCO 2010)
- "Snoopertext: A multiresolution system for text detection in complex visual scenes", R. Minetto, N. Thome, M. Cord, J. Fabrizio and B. Marcotegui (ICIP 10).
- "Adquisición de entornos tridimensionales empleando un Captor laser bidimensional", Luis Antonio García, Flavio Augusto Prieto (CONCAPAN XXXI-2011).
- "Automatische Bestimmung von Faserradienverteilungen", Hellen Altendorf, Stephan Didas und Till Batt (Forum Bildverarbeitung 2010)
- "Text Detection and Recognition in Urban Scenes", R. Minetto, N. Thome, M. Cord, J. Stolfi, F. Pr´ecioso, J. Guyomard, N.J Leite (CVRS workshop - ICCV 2011)
- "AdOtsu: An adaptive and parameterless generalization of Otsu's method for document image binarization", Reza Farrahi Moghaddam, Mohamed Cheriet (Pattern Recognition, 2012)
Bibtex
@inproceedings{fabrizio-icip09, author = {Fabrizio, J. and Marcotegui, B. and Cord, M.}, title = {Text segmentation in natural scenes using toggle-mapping}, booktitle = {Proceedings of the 16th IEEE International Conference on Image Processing}, date = {November 7-10}, location = {Cairo, Egypt}, pages = {2349--2352}, year = {2009}, }
City Models, Roads and Traffic (ISPRS Workshop - CMRT09). September 3-4, 2009. Paris, France
Abstract :
We offer in this article, a method for text extraction in images issued from city scenes. This method is used in the French iTowns project (iTowns ANR project, 2008) to automatically enhance cartographic database by extracting text from geolocalized pictures of town streets. This task is difficult as 1. text in this environment varies in shape, size, color, orientation... 2. pictures may be blurred, as they are taken from a moving vehicle, and text may have perspective deformations, 3. all pictures are taken outside with various objects that can lead to false positives and in unconstrained conditions (especially light varies from one picture to the other). Then, we can not make the assumption on searched text. The only supposition is that text is not handwritten. Our process is based on two main steps: a new segmentation method based on morphological operator and a classification step based on a combination of multiple SVM classifiers. The description of our process is given in this article. The efficiency of each step is measured and the global scheme is illustrated on an example.
Citations : 6
Bibtex
[+] "Fast Implementation of the Ultimate Opening", J. Fabrizio and B. MarcoteguiWe offer in this article, a method for text extraction in images issued from city scenes. This method is used in the French iTowns project (iTowns ANR project, 2008) to automatically enhance cartographic database by extracting text from geolocalized pictures of town streets. This task is difficult as 1. text in this environment varies in shape, size, color, orientation... 2. pictures may be blurred, as they are taken from a moving vehicle, and text may have perspective deformations, 3. all pictures are taken outside with various objects that can lead to false positives and in unconstrained conditions (especially light varies from one picture to the other). Then, we can not make the assumption on searched text. The only supposition is that text is not handwritten. Our process is based on two main steps: a new segmentation method based on morphological operator and a classification step based on a combination of multiple SVM classifiers. The description of our process is given in this article. The efficiency of each step is measured and the global scheme is illustrated on an example.
Citations : 6
- "Détection de texte dans des images couleur", Stefania Calarasanu - rapport sous la direction de S. Dubuisson - UPMC
- "A Novel Method for Efficient Text Extraction from Real Time Images with Diversified Background using Haar Discrete Wavelet Transform and K-Means Clustering", Narasimha Murthy K N, Y S Kumaraswamy (IJCSI 11)
- "Robust Text Detection in Natural Images with Edge-enhanced Maximally Stable Extremal Regions", H. Chen, S. S. Tsai, G. Scroth, D. Chen, R. Grzeszczuk, and B. Girod (ICIP 2011)
- "A Novel Algorithm for Text Detection and Localization in Natural Scene Images", Sezer Karaoglu, Basura Fernando, Alain Trémeau (DICTA 2010)
- "Snoopertext: A multiresolution system for text detection in complex visual scenes", R. Minetto, N. Thome, M. Cord, J. Fabrizio and B. Marcotegui (ICIP 10).
- "Text Detection and Recognition in Urban Scenes", R. Minetto, N. Thome, M. Cord, J. Stolfi, F. Pr´ecioso, J. Guyomard, N.J Leite (CVRS workshop - ICCV 2011)
Bibtex
@inproceedings{fabrizio-cmrt09, author = {J. Fabrizio and M. Cord and B. Marcotegui}, title = {Text Extraction from Street Level Images}, booktitle = {City Models, Roads and Traffic (ISPRS Workshop - CMRT09), date = {September 3-4}, pages = {}, address = {Paris, France}, year = {2009}, }
International Symposium on Mathematical Morphology (ISMM'09). Groningen, The Netherlands, August 24-27, 2009.
Abstract :
We present an efficient implementation of the ultimate attribute opening operator. In this implementation, the ultimate opening is computed by processing the image maxtree representation. To show the efficiency of this implementation, execution time is given for various images at different scales. A quasi-linear dependency with the number of pixels is observed. This new implementation makes the ultimate attribute opening usable in real time. Moreover, the use of the maxtree allows us to process specific zones of the image independently, with a negligible additional computation time.
Citations : 9
Bibtex
[+] "Ouverture Ultime : un outil pour la segmentation. Application à la localisation de texte.", J. Fabrizio, B. MarcoteguiWe present an efficient implementation of the ultimate attribute opening operator. In this implementation, the ultimate opening is computed by processing the image maxtree representation. To show the efficiency of this implementation, execution time is given for various images at different scales. A quasi-linear dependency with the number of pixels is observed. This new implementation makes the ultimate attribute opening usable in real time. Moreover, the use of the maxtree allows us to process specific zones of the image independently, with a negligible additional computation time.
Citations : 9
- "Ultimate Opening and Gradual Transitions", Beatriz Marcotegui, Jorge Hernández, and Thomas Retornaz (ISMM 2011)
- "Analyse morphologique d'images pour la modélisation d'environnements urbains", Jorge Eduardo Hernández Londoño (PhD Thesis 2009)
- "Ultimate Attribute Opening Segmentation with Shape Information", Jorge Hernández and Beatriz Marcotegui (ISMM 2009)
- "Shape ultimate attribute opening", Jorge Hernández, Beatriz Marcotegui (Image and Vision Computing 2011)
- "Fast and efficient FPGA implementation of connected operators", Nicolas Ngan, Eva Dokladalova, Mohamed Akil, Fançois Contou-Carrere (Journal of Systems Architecture 2011)
- "Text detection in street level images", Jonathan Fabrizio, Beatriz Marcotegui, Matthieu Cord (Pattern Analysis Applications, 2013)
- "Extraction of numerical residues in families of levelings", Alves, W.A.L. ; Morimitsu, A. ; Sanchez Castro, J. ; Fumio Hashimoto, R.(Graphics, Patterns and Images - SIBGRAPI - 2013)
- "ULTIMATE GRAIN FILTER", Wonder Alexandre Luz Alves and Ronaldo Fumio Hashimoto, (ICIP 2014)
- "RESIDUAL APPROACH ON A HIERARCHICAL SEGMENTATION", Beatriz Marcotegui, (ICIP 2014)
Bibtex
@inproceedings{fabrizio-ismm09, author = {Fabrizio, J. and Marcotegui, B.}, title = {Fast Implementation of the Ultimate Opening}, booktitle = {International Symposium on Mathematical Morphology (ISMM'09)}, date = {August 24-27, 2009}, pages = {272--281}, address = {Groningen, The Netherlands}, year = {2009}, }
31th meeting of the International Society for Stereology (ISS08), French section, Paris, Ecole des Mines, 7 february 2008
Abstract :
Nous présentons ici un système de segmentation des images basé sur l'opérateur d'ouverture ultime en vue d'extraire les caractères potentiels de l'image. Cet opérateur semble efficace pour cette tâche, mais souffre toutefois de quelques limitations que nous identifions. Pour repousser ces limites, notre système analyse le découpage proposé par l'ouvert ultime et le corrige en fusionnant certaines régions connexes et en fragmen- tant d'autres régions. Les résultats sont illustrés et le gain apporté par la correction est quantifié.
Citations : 0
Bibtex
[+] "Utilisation des distances tangentes pour la compensation de mouvement", J. Fabrizio and S. DubuissonNous présentons ici un système de segmentation des images basé sur l'opérateur d'ouverture ultime en vue d'extraire les caractères potentiels de l'image. Cet opérateur semble efficace pour cette tâche, mais souffre toutefois de quelques limitations que nous identifions. Pour repousser ces limites, notre système analyse le découpage proposé par l'ouvert ultime et le corrige en fusionnant certaines régions connexes et en fragmen- tant d'autres régions. Les résultats sont illustrés et le gain apporté par la correction est quantifié.
Citations : 0
Bibtex
@misc{fabrizio-iss08, author = {Fabrizio, J. and Marcotegui B.}, title = {Ouverture Ultime : un outil pour la segmentation. Application à la localisation de texte.}, booktitle = {31th meeting of the International Society for Stereology (ISS08)}, date = {february 2008}, address = {Paris, France}, year = {2008}, url = {http://jo.fabrizio.free.fr/recherche/fabmarc_iss08.pdf} }
COmpression et REprésentation des Signaux Audiovisuels (CORESA'07). Montpellier, France, Nov. 8-9 2007.
Abstract :
Dans cet article, nous présentons un algorithme de compensation de mouvement qui repose sur l'utilisation des distances tangentes. Contrairement aux algorithmes de Block-Matching classiques, qui prédisent uniquement l'évolution de la position spatiale de blocs image, notre algorithme prédit aussi l'évolution des pixels au cours du temps à l'intérieur même de ces blocs. De ce fait, l'erreur de prédiction diminue fortement. Intégré dans une chaîne de compression complète, notre algorithme pourrait en améliorer le taux de compression et la qualité de reconstruction. Le nombre d'images de référence nécessaires à l'encodage d'une séquence peut aussi être diminué par l'utilisation de notre algorithme.
Citations : 0
Bibtex
[+] "Motion estimation using tangent distance", J. Fabrizio and S. DubuissonDans cet article, nous présentons un algorithme de compensation de mouvement qui repose sur l'utilisation des distances tangentes. Contrairement aux algorithmes de Block-Matching classiques, qui prédisent uniquement l'évolution de la position spatiale de blocs image, notre algorithme prédit aussi l'évolution des pixels au cours du temps à l'intérieur même de ces blocs. De ce fait, l'erreur de prédiction diminue fortement. Intégré dans une chaîne de compression complète, notre algorithme pourrait en améliorer le taux de compression et la qualité de reconstruction. Le nombre d'images de référence nécessaires à l'encodage d'une séquence peut aussi être diminué par l'utilisation de notre algorithme.
Citations : 0
Bibtex
@inproceedings{fabrizio-coresa07, author = {Fabrizio, J. and Dubuisson, S.}, title = {Utilisation des distances tangentes pour la compensation de mouvement}, booktitle = {12èmes journées - COmpression et REprésentation des Signaux Audiovisuels (CORESA 2007)}, date = {November 8-9, 2007}, pages = {95-99}, address = {Montpellier, France}, year = {2007}, }
International Conference of Image Processing (ICIP'07). San Antonio (Texas), USA, Oct. 16-19 2007.
Abstract :
In this paper, we present a method based on tangent distance to estimatemotion in image sequences. Tangent distance combines an intuitive understanding and effective modeling of differences between patterns. This tool was first introduced and successfully applied in character recognition. It allows to compare patterns according to small transformations (translations, rotations, etc.). We show, how to take advantages of some properties of tangent distances to perform a robust motion estimation algorithm. Particularly, the presented algorithm can easily be adapted and optimized to various types of movements and can also be used to estimate optical flow in image sequences. Moreover, and despite a time of computation a bit long, this algorithm can be massively paralleled.
Citations : 0
[+] "The Perspective-N-Point Problem for Catadioptric Sensors: An Analytical Approach", J. Fabrizio, J. Devars,In this paper, we present a method based on tangent distance to estimatemotion in image sequences. Tangent distance combines an intuitive understanding and effective modeling of differences between patterns. This tool was first introduced and successfully applied in character recognition. It allows to compare patterns according to small transformations (translations, rotations, etc.). We show, how to take advantages of some properties of tangent distances to perform a robust motion estimation algorithm. Particularly, the presented algorithm can easily be adapted and optimized to various types of movements and can also be used to estimate optical flow in image sequences. Moreover, and despite a time of computation a bit long, this algorithm can be massively paralleled.
Citations : 0
@inproceedings{-, author = {J. Fabrizio and S. Dubuisson}, title = {Motion estimation using tangent distance}, booktitle = {Proceedings of the International Conference on Image Processing}, date = {September 16-19}, pages = {489-492}, address = {San Antonio, Texas, USA}, year = {2007}, }
International Conference on Computer Vision and Graphics (ICCVG'04), warsaw, september 2004
Published by Kluwer in the book series COMPUTATIONAL IMAGING AND VISION.
Abstract :
The perspective-N-point problem is a well known issue in computer vision. It consists in the determination of the distance between the camera and a set of points well known in an object coordinate space. This problem has been extensively treated in the literature and is still opened. Many solutions already exist. All these approaches consider only common planar camera. We propose, with a new formulation, to extend this problem to non linear imaging sensors: catadioptric panoramic sensors. The proposed approach permits to get a strictly analytical solution to the perspective-N-point problem usable with this kind of sensors.
Citations : 4
Bibtex
[+] "Une solution analytique au problème « Perspective-N-Points » et son extension aux capteurs catadioptriques", J. Fabrizio, J. DevarsThe perspective-N-point problem is a well known issue in computer vision. It consists in the determination of the distance between the camera and a set of points well known in an object coordinate space. This problem has been extensively treated in the literature and is still opened. Many solutions already exist. All these approaches consider only common planar camera. We propose, with a new formulation, to extend this problem to non linear imaging sensors: catadioptric panoramic sensors. The proposed approach permits to get a strictly analytical solution to the perspective-N-point problem usable with this kind of sensors.
Citations : 4
- "Noncentral catadioptric systems with quadric mirrors - geometry and calibration -", Nuno Miguel Mendonca Da Silva Goncalves (PhD Thesis 2008)
- "Solution of the Perspective-Three-Point Problem", Loic Merckel and Toyoaki Nishida (IEA/AIE 2007)
- "Stabilization and location of a four rotor helicopter applying vision", Hugo ROMERO, Ryad BENOSMAN, Rogelio LOZANO, Proc. American Control Conference, 2006
- "Estimation numérique des courbes épipolaires pour les capteurs omnidirectionnels", J. Fabrizio, J. Devars, revue Traitement du Signal (TS 2005).
Bibtex
@inproceedings{fabrizio-iccvg04, author = {Fabrizio, J. and Devars, J.}, title = {The Perspective-N-Point Problem for Catadioptric Sensors: An Analytical Approach}, booktitle = {International Conference on Computer Vision and Graphics (ICCVG'04)}, date = {september 22-24, 2004}, pages = {599-606}, address = {Warsaw, Poland}, year = {2004}, url = {http://jo.fabrizio.free.fr/recherche/fabdev_iccvg04.pdf} }
19th GRETSI Symposium on Signal and Image Processing (GRETSI'03), Paris, 8 - 11 September 2003.
Abstract :
The problem of estimating the distance between a unique image sensor and a well known pattern composed by N points (The Perspective-N-Points problem) has been extensively treated and many solutions exist. By the use of a new set of equations, we get, whereas most of existing solutions, a strictly analytical solution. This solution first request few and constant CPU time which allow us to use it in real time applications and second is usable with nonlinear sensor (particularly omnidirectional sensor).
Citations : 0
Bibtex
[+] "Calibration of Panoramic Catadioptric Sensors Made Easier", J. Fabrizio, J.-P. Tarel, et R. Benosman,The problem of estimating the distance between a unique image sensor and a well known pattern composed by N points (The Perspective-N-Points problem) has been extensively treated and many solutions exist. By the use of a new set of equations, we get, whereas most of existing solutions, a strictly analytical solution. This solution first request few and constant CPU time which allow us to use it in real time applications and second is usable with nonlinear sensor (particularly omnidirectional sensor).
Citations : 0
Bibtex
@inproceedings{fabrizio-gretsi03, author = {Fabrizio, J. and Devars, J.}, title = {Une solution analytique au problème « Perspective-N-Points » et son extension aux capteurs catadioptriques}, booktitle = {19th GRETSI Symposium on Signal and Image Processing (GRETSI'03)}, date = {September 8-11, 2003}, pages = {??-??}, address = {Paris, France}, year = {2003}, url = {http://jo.fabrizio.free.fr/recherche/fabdev_gretsi03.pdf} }
IEEE European Conference Computer Vision, Workshop on Omnidirectional Vision (ECCV Workshop-OMNIVIS 02), page 45-52, Copenhague, 02 Juin 2002.
Abstract :
We present a new method to calibrate panoramic catadioptric sensors. While many methods exist for planar cameras, it is not the case for panoramic catadioptric sensors. The aim of the proposed calibration is not to estimate the mirror surface parameters which can be known very accurately, but to estimate the intrinsic parameters of the CCD camera and the pose parameters of the CCD camera with respect to the mirror. Unless a telecentric lens is used, this pose must be estimated, particularly for sensors that have a unique effective view point. The developed method is based on the original and simple idea that the mirror external and internal boundaries can be used as a 3D calibration pattern. The improvement introduced by our approach is demonstrated on synthetic experiments with incorrectly aligned sensors and validation tests on real images are described. The proposed technique opens new ways for better designed catadioptric sensors where self-calibration can be easily performed in real-time in a completely autonomous way. In particular this should allow to avoid errors due to vibrations one can notice when using catadioptric sensors in practical situations.
Citations : 64
Bibtex
We present a new method to calibrate panoramic catadioptric sensors. While many methods exist for planar cameras, it is not the case for panoramic catadioptric sensors. The aim of the proposed calibration is not to estimate the mirror surface parameters which can be known very accurately, but to estimate the intrinsic parameters of the CCD camera and the pose parameters of the CCD camera with respect to the mirror. Unless a telecentric lens is used, this pose must be estimated, particularly for sensors that have a unique effective view point. The developed method is based on the original and simple idea that the mirror external and internal boundaries can be used as a 3D calibration pattern. The improvement introduced by our approach is demonstrated on synthetic experiments with incorrectly aligned sensors and validation tests on real images are described. The proposed technique opens new ways for better designed catadioptric sensors where self-calibration can be easily performed in real-time in a completely autonomous way. In particular this should allow to avoid errors due to vibrations one can notice when using catadioptric sensors in practical situations.
Citations : 64
- "A convenient vision-based system for automatic detection of parking spaces in indoor parking lots using wide-angle cameras", SE Shih, WH Tsai (TVT 2014)
- "Calibration of mirror position and extrinsic parameters in axial non-central catadioptric systems", L Perdigoto and H Araujo (CVIU 2013)
- "The camera itself as a calibration pattern: A novel self-calibration method for non-central catadioptric cameras", Z Xiang, B Sun and X Dai (Sensors 2012)
- "An FPGA-based omnidirectional vision sensor for motion detection on mobile robots", Jones Y. Mori, Janier Arias-Garcia, Camilo Sánchez-Ferreira, Daniel M. Muñoz, Carlos H. Llanos, J. M. S. T. Motta (IJRC 2012)
- "A New Two-Omni-Camera System with a Console Table for Versatile 3D Vision Applications and Its Automatic Adaptation to Imprecise Camera Setups", Shen-En Shih and Wen-Hsiang Tsai (Proceedings of the 17th international conference on Advances in multimedia modeling MMM 2011)
- "Omni-Directional Camera and Fuzzy Logic Path Planner for Autonomous Sailboat Navigation", Romero Ramirez, M.A. and Guo, Y. and Ieng, S.H. and Plumet, F. and Benosman, R and Gas, B. (Research in Computing Science, Special Issue in Advances in Computer Science and Electronic Systems 2011)
- "Camera Models and Fundamental Concepts Used in Geometric Computer Vision", Peter Sturm, Srikumar Ramalingam, Jean-Philippe Tardif, Simone Gasparini and João Barreto (FTCGV 2011)
- "Reactive path planning for autonomous sailboat using an omni-directional camera for obstacle detection", Y Guo, M Romero, SH Ieng and F Plumet (ICM 2011)
- Non-Central Catadioptric Sensors Auto-Calibration", Abd El Rahman Shabayek (Phd Thesis 2009)
- "Analytic image unwarping by a systematic calibration method for omni-directional cameras with hyperbolic-shaped mirrors", Sheng-Wen Jeng and Wen-Hsiang Tsai, (Image Vision Computing 2008)
- "Noncentral catadioptric systems with quadric mirrors - geometry and calibration -", Nuno Miguel Mendonca Da Silva Goncalves (PhD Thesis 2008)
- "Stereo Matching and 3D Reconstruction via an Omnidirectional Stereo Sensor", Lei He, Chuanjiang Luo, Feng Zhu and Yingming Hao (ISBN 978-953-7619-01-5 2008)
- "Measurement of three-dimensional mirror parameters by polarization imaging applied to catadioptric camera calibration", Olivier Morel, Ralph Seulin, David Fofi (Journal of Electric Imaging 2008)
- "An analytical solution to the perspective-n-point problem for common planar camera and for catadioptric sensor", Jonathan Fabrizio, Jean Devars (IJIG 2008)
- "Autonomous Agents in a Dynamic Collaborative Environment", Gourab Sen Gupta (PhD Thesis 2008)
- "Méthodologies d'estimation et de commande à partir d'un système de vision", Ezio Malis (HRD 2008)
- ?"Processing Sparse Panoramic Images via Space Variant Operators", Sonya Coleman, Bryan Scotney, Dermot Kerr (Journal of Mathematical Imaging and Vision 2008)
- "A Novel Omnidirectional Stereo Vision System via a Single Camera", Chuanjiang Luo, Liancheng Su and Feng Zhu from book "Scene Reconstruction, Pose Estimation and Tracking" - ISBN 978-3-902613-06-6 - Edited by: Rustam Stolkin - Publisher: I-Tech Education and Publishing, Vienna, Austria - June 2007
- "Low-cost method for the estimation of the shape of quadric mirrors and calibration of catadioptric cameras", Nuno Gonçalves and Helder Araújo (Optical Engineering 2007)
- "Single-Image Calibration of Off-Axis Catadioptric Cameras Using Lines", Aglioti Vincenzo, Taddei Pierluigi, Boracchi Giacomo, Gasparini Simone and Giusti Alessandro (OMNIVIS 2007)
- "Real-Time Visual Servoing Control of a Four-Rotor Rotorcraft", Romero Hugo, Salazar Sergio, Lozano Rogelio, Benosman Ryad (ALCOSP'07)
- "Calibration of a Panoramic Stereovision Sensor : Analytical vs Interpolation-Based Methods", Ragot, N., Ertaud, J-Y., Savatier, X., Mazari, B. (IEEE 32nd Annual Conference on Industrial Electronics IECON 2006)
- "Stabilization and location of a four rotor helicopter applying vision", Hugo ROMERO, Ryad BENOSMAN, Rogelio LOZANO (Proc. American Control Conference 2006)*
- "Calibration Method for Misaligned Catadioptric Camera", Tomohiro Mashita, Yoshio Iwai, Masahiko Yachida (IEICE Transactions on Information and Systems 2006)
- "Calibrage non biaisé d'un capteur central catadioptrique - Unbiased central catadioptric camera calibration", Christopher Mei, Patrick Rives (RFIA 2006)
- "Geometric Construction of the Caustic Surface of Catadioptric Non-Central Sensors", Sio-hoi Ieng, Ryad Benosman (Image Beyong the Pinhole Camera - 2006)
- "Calibration of a Panoramic Stereovision Sensor : Analytical vs Interpolation-Based Methods", Ragot, N., Ertaud, J-Y., Savatier, X., Mazari, B. (IEEE 32nd Annual Conference on Industrial Electronics, IECON 2006)
- "Calibration of panoramic catadioptric sensors made easier", , 3DVG meeting - CCU Vision Lab (tw), (2/14/06)
- "Generic imaging models: Calibration and 3d reconstruction algorithms", S Ramalingam, (PhD Thesis 2006)
- "Parameter Extraction for Hyperboloidal Catadioptric Omnidirectional Cameras (in Turkish)", Bastanlar, Y., Yardimci, Y., (IEEE National Conference on Signal Processing and Applications, - SIU 2005)
- "Etalonnage de caméras catadioptriques hyperboloïdes", F. Comby, O. Strauss, C. Caderas de Kerleau (TS 2005)
- "Les surfaces caustiques par la géométrie - Application aux capteurs catadioptriques", I. Sio-Hoï, R. Benosman (TS 2005)
- "Vision Omnidirectionnelle et Robotique - Rapport Final", Projet "Jeunes chercheur" P13 du GdR Isis (2005)
- "Fast Dense Panoramic Stereovision", J.-J. Gonzalez-Barbosa, S. Lacroix, Int. Conf. on Robotics and Automation (ICRA 2005)
- "Stéréovision panoramique dense", S. Lacroix, J.-J. Gonzalez-Barbosa, (TS 2005)
- "La Vision Omnidirectionnelle", El Mustapha Mouaddib, Journées Nationales de la Recherche en Robotique (JNNR 2005)
- "Introduction à la Vision Panoramique Catadioptrique", El Mustapha Mouaddib, (TS 2005)
- "Calibration Method for Misaligned Catadioptric Camera", Tomohiro Mashita, Yoshio Iwai, Masahiko Yachida (OMNIVIS 2005)
- "Estimation numérique des courbes épipolaires pour les capteurs omnidirectionnels", J. Fabrizio, J. Devars, (TS 2005)
- "Etude des proprietes geometriques des capteurs panoramiques", Sio-Hoi IENG (PhD Thesis 2005)*
- "Depth Perception with a Single Camera", Jonathan R. Seal, Donald G. Bailey, Gourab Sen Gupta, International Conference on Sensing Technology (ICST 2005)
- "Parameter extraction and image enhancement for catadioptric omnidirectional cameras", Yalin Bastanlar, (PhD Thesis, METU 2005)
- "H.264 Based Coding of Omnidirectional Video", Ingo Bauermann, Matthias Mielke and Eckehard Steinbach (ICCVG 04)
- "3D Metric Reconstruction from Uncalibrated Omnidirectional Images", Branislav Micusik, Daniel Martinec and Tomas Pajdla (ACCV 2004)
- "Vision panoramique pour la robotique mobile : stereovision et localisation par indexation d'images", Jose-Joel Gonzalez-Barbosa, Phd Thesis, (2004)
- "Two-View Geometry of Omnidirectional Cameras", Branislav Mičušík, (PhD Thesis 2004)
- "Geometric construction of the caustic curves for catadioptric sensors", Sio-hoi Ieng, Benosman (ICIP 04)
- "The perspective-N-point problem for catadioptric sensors: an analytical approach", Jonathan Fabrizio, Jean Devars (ICCVG 04)
- "Para-catadioptric camera auto-calibration from epipolar geometry", Branislav Micusik and Tomas Pajdla (ACCV 2004)
- "An Ellipses Correspondence Based Self-Calibration Algorithm for Omnidirectional Sensors", M. M. Rahman and S. Horiguchi, Proc. International Conference on Computer and Information Technology (ICCIT 2004)*
- "Utilisation d'un compas visuel pour la navigation d'un robot mobile", Stéphane Gourichon, (PhD Thesis 2004)
- "Mapping and Localization from a Panoramic Vision Sensor", Roland Bunschoten (Phd Thesis 2003)
- "Un algorithme rapide de stéréovision panoramique dense", J-J. Gonzalez-Barbosa and S. Lacroix. (Technical report, LAAS/CNRS. (In french) 2003)*
- "Omnidirectional Vision System for Mobile Robot Localization in the RoboCup Environment", Juregen wolf (Diplomarbeit 2003)*
- "Particle Filter for Self Localization using Panoramic Vision", Jurgen Wolf and Axel Pinz (In Proc. of 26th Workshop of the Austrian Association for Pattern Recognition ÖAGM/AAPR 2003)
- "Precise Image Unwarping of Omnidirectional Cameras with Hyperbolic-Shaped Mirrors", Sheng-Wen Jeng and Wen-Hsiang Tsai (16th IPPR Conference on Computer Vision, Graphics and Image Processing CVGIP 03)
- "Une solution analytique au problème « Perspective-N-Points » et son extension aux capteurs catadioptriques", Jonathan Fabrizio, Jean Devars (GRETSI 03)
- "An Efficient Dynamic Multi-Angular Feature Points Matcher for Catadioptric Views", Sio-hoï Ieng, Ryad Benosman, Jean Devars (OMNIVIS 03)
- "Perception par caméra des bords de route", Jean-Philippe Tarel, Book, Collections Etudes et Recherches des Laboratoires des Ponts et Chaussées (LCPC 03)*
- "Estimating ego-motion using a panoramic sensor: Comparison between a bio-inspired and a camera-calibrated method", S. Gourichon, J.A. Meyer, S. H. Iengy, L.Smadja, R. Benosman, (Symposium on Biologically Inspired Vision, Theory and Application AISB 03)
- "Generation d'environnements 3D denses à partir d'images panoramiques cylindrinques", Laurent Smadja (PhD Thesis 2003)
- "Vision Omnidirectionnelle" Cédric Demonceaux
- "Omnidirectional Vision - An overview", Pascal Vasseur
- "Capture and Calibration of Omnidirectional Cameras", Miguel Morgado*
Bibtex
@inproceedings{fabrizio-omnivis02, author = {Fabrizio, J. and Tarel, J.-P. and Benosman, R.}, title = {Calibration of Panoramic Catadioptric Sensors Made Easier}, booktitle = {Proceedings of IEEE Workshop on Omnidirectional Vision (Omnivis'02)}, date = {June 2nd}, pages = {45-52}, address = {Copenhagen, Denmark}, year = {2002}, url = {http://jo.fabrizio.free.fr/recherche/fabrizio_omnivis02.pdf} }
Oral Presentations
"Utilisation des distances tangentes pour la compensation de mouvement : Application au codec Theora", J. FabrizioSéminaire Perfromance et Généricité EPITA Research and Development Laboratory, Le Kremlin-Bicêtre, France, 18 mai 2011
"L'algorithme « ICP »", J. Fabrizio
Séminaire mouvement, Centre de Morphologie Mathematique, Ecole des Mines de Paris, Fontainebleau, 19 mars 2008.
"Détection et localisation d'obstacles et apports des capteurs catadioptriques pour la navigation", J. Fabrizio, O. Kloc, J. Devars,
Journées de synthèse ARCOS, Arcueil 25 Septembre 2003
"Détection d'obstacles coopératifs", J. Fabrizio, O. Kloc, J. Devars,
Journées de synthè, Satory, 19 juin 2003
"Solution analytique du « PNP » étendue aux capteurs cartadioptriques", J. Fabrizio, J. Devars,
Journée Vision Panoramique, GT5 Vision - GDR ISIS, Paris, 5 juin 2003.
Technical documents
How to interpret RAW Bayer data from the PiCam HQ, Jonathan FabrizioTechnical document 06/20