Cyril Charron - Academia.edu (original) (raw)
Papers by Cyril Charron
2010 IEEE International Conference on Image Processing, 2010
... 1–8, 2008. [4] J. Shi and J. Malik, “Motion segmentation and tracking us-ing normalized cut,”... more ... 1–8, 2008. [4] J. Shi and J. Malik, “Motion segmentation and tracking us-ing normalized cut,” in International Conference on Computer Vision, 1998, pp. 1154–1160. ... 28, no. 3, pp. 463–468, March 2006. [9] C. Charron, Y. Hicks, and P. Hall, “Applying incremental learn-ing to ...
Lecture Notes in Computer Science, 2006
... 2 Omni-Appearance Based Localization Algorithm ... Kernel function f : MS ↦→ Cf defines a map... more ... 2 Omni-Appearance Based Localization Algorithm ... Kernel function f : MS ↦→ Cf defines a mapping from spherical image MS to feature space Cf ... It relies on a Fourier decomposition along the lines of the 3 fi could be seen as the probability for a characteristic to belong to a given ...
Lecture Notes in Computer Science, 2006
... Kernel function f : MS ↦→ Cf defines a mapping from spherical image MS to feature space Cf . ... more ... Kernel function f : MS ↦→ Cf defines a mapping from spherical image MS to feature space Cf . ... 3 fi could be seen as the probability for a characteristic to belong to a ... 4. I. Ulrich and I. Nourbakhsh, Appearance-based place recognition for topologi-cal localization, in International ...
2005 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2005
The present study proposes an innovative approach to qualitative mobile robot's l... more The present study proposes an innovative approach to qualitative mobile robot's localization using the concept of integral invariant on omnidirectional images. They are invariant depending on the image transformations caused by the movements of the robot. Several methods have been suggested to construct such invariants but they often rely on hypotheses about the transformation group which do not hold any
Journal of Vision, 2010
Abstract Traditional accounts of adaptation to a rotation of the optic array (due to prisms) iden... more Abstract Traditional accounts of adaptation to a rotation of the optic array (due to prisms) identify both visual and proprioceptive sites (Redding & Wallace, 1985). Last year (Brandwood, Rushton & Charron, VSS 2009), we reported the results of a walking ...
Journal of Vision, 2010
When an error is introduced into the mapping between visual direction and the movement of the fee... more When an error is introduced into the mapping between visual direction and the movement of the feet, an observer will initially take a curved path when walking to a stationary visual target. After a period of walking, remapping, or adaptation, will occur resulting in a ...
Autonomous Robots, 2011
This paper describes a method for spatial representation, place recognition and qualitative self-... more This paper describes a method for spatial representation, place recognition and qualitative self-localization in dynamic indoor environments, based on omnidirectional images. This is a difficult problem because of the perceptual ambiguity of the acquired images, and their weak robustness to noise, geometrical and photometric variations of real world scenes. The spatial representation is built up invariant signatures using Invariance Theory where we suggest to adapt Haar invariant integrals to the particular geometry and image transformations of catadioptric omnidirectional sensors. It follows that combining simple image features in a process of integration over visual transformations and robot motion, can build discriminant percepts about robot spatial locations. We further analyze the invariance properties of the signatures and the apparent relation between their similarity measures and metric distances. The invariance properties of the signatures can be adapted to infer a hierarchical process, from global room recognition to local and coarse robot localization.
Dans cetteétude, nous nous intéressonsà la localisation d'un robot mobile par indexation d'images... more Dans cetteétude, nous nous intéressonsà la localisation d'un robot mobile par indexation d'images omnidirectionnelles. La localisation est qualitative et le robot doit déterminer la pièce dans laquelle il se trouve. L'indexation est réaliséeà partir d'invariants intégraux calculés sur les images omnidirectionnelles. Notre méthode utilise l'intégrale de Haar dans laquelle nous intégrons un modèle du capteur omnidirectionnel. L'approche intégrale nous permet de prendre en compte simplement la géométrie du capteur ainsi que les déplacements du robot pour le calcul des invariants. La robustesse de cette nouvelle méthodè a des déplacements dans une pièce est supérieureà une méthode basée sur l'histogramme.
2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops, 2009
... Thus, they use the MML for-malism to derive a modified update step. It results in a mod-ified... more ... Thus, they use the MML for-malism to derive a modified update step. It results in a mod-ified update of the priors: ... However, to avoid a costly iterative procedure, we replace the Bhat-tacharyya distance in Eq. (7) by the Kullback-Leibler diver-gence and obtain Eq. ...
Procedings of the British Machine Vision Conference 2009, 2009
Active Appearance Models (AAM) are a useful and popular tool for modelling facial variations. The... more Active Appearance Models (AAM) are a useful and popular tool for modelling facial variations. They have been used in face tracking, recognition and synthesis applications. For modelling facial dynamics of speech, they have been used in conjunction with Hidden Markov Models (HMM). However, the high dimensionality of the training data and of the resulting AAMs leads to long learning time of HMMs and thus imposes serious limitations on their joint use.
2011 IEEE International Conference on Computer Vision Workshops (ICCV Workshops), 2011
ABSTRACT The challenge addressed in this paper is the classification of visual objects by robots.... more ABSTRACT The challenge addressed in this paper is the classification of visual objects by robots. Visual classification is an active field within Computer Vision, with excellent results achieved recently.
2010 IEEE International Conference on Image Processing, 2010
... 1–8, 2008. [4] J. Shi and J. Malik, “Motion segmentation and tracking us-ing normalized cut,”... more ... 1–8, 2008. [4] J. Shi and J. Malik, “Motion segmentation and tracking us-ing normalized cut,” in International Conference on Computer Vision, 1998, pp. 1154–1160. ... 28, no. 3, pp. 463–468, March 2006. [9] C. Charron, Y. Hicks, and P. Hall, “Applying incremental learn-ing to ...
Lecture Notes in Computer Science, 2006
... 2 Omni-Appearance Based Localization Algorithm ... Kernel function f : MS ↦→ Cf defines a map... more ... 2 Omni-Appearance Based Localization Algorithm ... Kernel function f : MS ↦→ Cf defines a mapping from spherical image MS to feature space Cf ... It relies on a Fourier decomposition along the lines of the 3 fi could be seen as the probability for a characteristic to belong to a given ...
Lecture Notes in Computer Science, 2006
... Kernel function f : MS ↦→ Cf defines a mapping from spherical image MS to feature space Cf . ... more ... Kernel function f : MS ↦→ Cf defines a mapping from spherical image MS to feature space Cf . ... 3 fi could be seen as the probability for a characteristic to belong to a ... 4. I. Ulrich and I. Nourbakhsh, Appearance-based place recognition for topologi-cal localization, in International ...
2005 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2005
The present study proposes an innovative approach to qualitative mobile robot's l... more The present study proposes an innovative approach to qualitative mobile robot's localization using the concept of integral invariant on omnidirectional images. They are invariant depending on the image transformations caused by the movements of the robot. Several methods have been suggested to construct such invariants but they often rely on hypotheses about the transformation group which do not hold any
Journal of Vision, 2010
Abstract Traditional accounts of adaptation to a rotation of the optic array (due to prisms) iden... more Abstract Traditional accounts of adaptation to a rotation of the optic array (due to prisms) identify both visual and proprioceptive sites (Redding & Wallace, 1985). Last year (Brandwood, Rushton & Charron, VSS 2009), we reported the results of a walking ...
Journal of Vision, 2010
When an error is introduced into the mapping between visual direction and the movement of the fee... more When an error is introduced into the mapping between visual direction and the movement of the feet, an observer will initially take a curved path when walking to a stationary visual target. After a period of walking, remapping, or adaptation, will occur resulting in a ...
Autonomous Robots, 2011
This paper describes a method for spatial representation, place recognition and qualitative self-... more This paper describes a method for spatial representation, place recognition and qualitative self-localization in dynamic indoor environments, based on omnidirectional images. This is a difficult problem because of the perceptual ambiguity of the acquired images, and their weak robustness to noise, geometrical and photometric variations of real world scenes. The spatial representation is built up invariant signatures using Invariance Theory where we suggest to adapt Haar invariant integrals to the particular geometry and image transformations of catadioptric omnidirectional sensors. It follows that combining simple image features in a process of integration over visual transformations and robot motion, can build discriminant percepts about robot spatial locations. We further analyze the invariance properties of the signatures and the apparent relation between their similarity measures and metric distances. The invariance properties of the signatures can be adapted to infer a hierarchical process, from global room recognition to local and coarse robot localization.
Dans cetteétude, nous nous intéressonsà la localisation d'un robot mobile par indexation d'images... more Dans cetteétude, nous nous intéressonsà la localisation d'un robot mobile par indexation d'images omnidirectionnelles. La localisation est qualitative et le robot doit déterminer la pièce dans laquelle il se trouve. L'indexation est réaliséeà partir d'invariants intégraux calculés sur les images omnidirectionnelles. Notre méthode utilise l'intégrale de Haar dans laquelle nous intégrons un modèle du capteur omnidirectionnel. L'approche intégrale nous permet de prendre en compte simplement la géométrie du capteur ainsi que les déplacements du robot pour le calcul des invariants. La robustesse de cette nouvelle méthodè a des déplacements dans une pièce est supérieureà une méthode basée sur l'histogramme.
2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops, 2009
... Thus, they use the MML for-malism to derive a modified update step. It results in a mod-ified... more ... Thus, they use the MML for-malism to derive a modified update step. It results in a mod-ified update of the priors: ... However, to avoid a costly iterative procedure, we replace the Bhat-tacharyya distance in Eq. (7) by the Kullback-Leibler diver-gence and obtain Eq. ...
Procedings of the British Machine Vision Conference 2009, 2009
Active Appearance Models (AAM) are a useful and popular tool for modelling facial variations. The... more Active Appearance Models (AAM) are a useful and popular tool for modelling facial variations. They have been used in face tracking, recognition and synthesis applications. For modelling facial dynamics of speech, they have been used in conjunction with Hidden Markov Models (HMM). However, the high dimensionality of the training data and of the resulting AAMs leads to long learning time of HMMs and thus imposes serious limitations on their joint use.
2011 IEEE International Conference on Computer Vision Workshops (ICCV Workshops), 2011
ABSTRACT The challenge addressed in this paper is the classification of visual objects by robots.... more ABSTRACT The challenge addressed in this paper is the classification of visual objects by robots. Visual classification is an active field within Computer Vision, with excellent results achieved recently.