### Citations

1012 | publications by the team in recent years - Major |

123 | Aggregating local image descriptors into compact codes
- Jégou, Perronnin, et al.
(Show Context)
Citation Context ...ctors on sparse or dense local SIFT features. The dense feature extraction was optimized, so that they can be computed in real time on video data. The implementation was used for several publications =-=[6]-=-, [16] and in our submisssion to the Trecvid 2012 MED task [31]. We provide a binary version of the local descriptor implementation, and the Fisher implementation is integrated in the Yael library, wi... |

117 | Dense trajectories and motion boundary descriptors for action recognition
- Wang, Kläser, et al.
- 2013
(Show Context)
Citation Context ... Cordelia Schmid [correspondant], Heng Wang.Project-Team LEAR 7 We have developed and made on-line available software for video description based on dense trajectories and motion boundary histograms =-=[28]-=-. The trajectories capture the local motion information of the video. A stateof-the-art optical flow algorithm enables a robust and efficient extraction of the dense trajectories. Descriptors are alig... |

56 | Weakly supervised learning of interactions between humans and objects
- Prest, Schmid, et al.
- 2011
(Show Context)
Citation Context ...large scale learning. In NIPS,2007. 1, 4 [9] K. Chatfield, V. Lempitsky, A. Vedaldi, and A. Zisserman. The devil is in the details: an evaluation of recent feature encoding methods. In BMVC,2011.1, 5 =-=[10]-=- K. Crammer and Y. Singer. On the algorithmic implementation of multiclass kernel-based vector machines. JMLR, 2001.1, 3 8 training. In CVPR,2011. 1, 2, 4, 7 [20] D. Lowe. Distinctive image features f... |

52 | Learning object class detectors from weakly annotated video
- Prest, Leistner, et al.
- 2012
(Show Context)
Citation Context ...icipants: Javier Civera, Vittorio Ferrari, Christian Leistner, Alessandro Prest, Cordelia Schmid. Object detectors are typically trained on a large set of still images annotated by bounding-boxes. In =-=[20]-=- we introduce an approach for learning object detectors from real-world web videos known only to contain objects of a target class. We propose a fully automatic pipeline that localizes objects in a se... |

50 | Towards good practice in large-scale learning for image classification
- Perronnin, Akata, et al.
- 2012
(Show Context)
Citation Context ...ithm used to train linear multiclass classifiers. It is biased towards large classification problems (many classes, many examples, high dimensional data). It can be used to reproduce the results from =-=[19]-=- on the ImageNet large scale classification challenge. It uses several optimization techniques, both algorithmic (scale factors to spare vector multiplications, vector compression with product quantiz... |

49 | A Component-wise EM Algorithm for Mixtures - Celeux, Chrétien, et al. |

41 | Hidden Markov random field model selection criteria based on mean field-like approximations - Forbes, Peyrard - 2003 |

31 | Discriminative spatial saliency for image classification
- Sharma, Jurie, et al.
(Show Context)
Citation Context ...to” based on the area around the arms i.e. ignoring the legs, and (ii) has intra class variations e.g. different readers may hold the books differently. Motivated by these observations, we propose in =-=[22]-=- to learn the discriminative spatial saliency of images while simultaneously learning a max-margin classifier for a given visual classification task. Using the saliency maps to weight the correspondin... |

30 | High dimensional discriminant analysis
- Bouveyron, Girard, et al.
- 2007
(Show Context)
Citation Context ... obtained by imposing to different groups to share the same values of some parameters This modelling yields a new supervised classification method called High Dimensional Discriminant Analysis (HDDA) =-=[4]-=-. Some versions of this method have been tested on the supervised classification of objects in images. This approach has been adapted to the unsupervised classification framework, and the related meth... |

28 |
Density measure for line-drawing simplification
- GRABLI, DURAND, et al.
- 2004
(Show Context)
Citation Context ...r set of operators. Freestyle was developed in the framework of a research project dedicated to the study of stylized line drawing rendering from 3D scenes. This research has lead to two publications =-=[23]-=-, [24]. Figure 2. Stylized plane using Freestyle. In 2008, Freestyle get a new life, completely outside Maverick or Inria: it was the basis of one of the 6 Google Summer of Code projects awarded to th... |

27 | Lifted coordinate descent for learning with trace-norm regularization
- Dudik, Harchaoui, et al.
- 2012
(Show Context)
Citation Context ...ing examples. 6.2.5. Lifted coordinate descent for learning with trace-norm regularization Participants: Miro Dudik [Microsoft Research], Zaid Harchaoui, Jérôme Malick [BiPoP Team Inria Grenoble]. In =-=[14]-=- we consider the minimization of a smooth loss with trace-norm regularization, which is a natural objective in multi-class and multi-task learning. Even though the problem is convex, existing approach... |

24 | procedures using mean field-like approximations for Markov model-based image segmentation - EM |

23 | Face recognition from caption-based supervision
- Guillaumin, Mensink, et al.
- 2012
(Show Context)
Citation Context ... Jordan, and J. D. McAuliffe. Convexity, classification, and risk bounds. In NIPS, 2003.7 [4] S. Bengio, J. Weston, and D. Grangier. Label embedding trees for large multi-class tasks. In NIPS, 2010.2 =-=[5]-=- A. Berg, J. Deng, and L. Fei-Fei. ILSVRC 2010. http://www.imagenet.org/challenges/LSVRC/2010/index. 1, 5 [6] A. Bordes, L. Bottou, P. Gallinari, and J. Weston. Solving multiclass support vector machi... |

23 | A Hill type estimator of the Weibull tail-coefficient
- Girard
- 2004
(Show Context)
Citation Context ... theory. For instance, in reliability, the distributions of interest are included in a semi-parametric family whose tails are decreasing exponentially fast. These so-called Weibull-tail distributions =-=[9]-=- are defined by their survival distribution function: P (X > x) = exp {−x θ ℓ(x)}, x > x0 > 0. (4) Gaussian, gamma, exponential and Weibull distributions, among others, are included in this family. An... |

21 |
Image categorization using Fisher kernels of non-i.i.d. image models
- Cinbis
- 2012
(Show Context)
Citation Context ...cy histograms. Implicitly, regions are assumed to be identically and independently distributed (iid), which is a very poor assumption from a modeling perspective; see Figure 5 for an illustration. In =-=[13]-=-, we introduce non-iid models by treating the parameters of bag-of-word models as latent variables which are integrated out, rendering all local regions dependent. Using the Fisher kernel we encode an... |

21 | E.: Interactive indirect illumination using voxel cone tracing. Computer Graphics Forum 30(7
- Crassin, Neyret, et al.
- 2011
(Show Context)
Citation Context ...n, data management and base rendering algorithm themself, we also worked on realtime light transport, and on quality prefiltering of complex data. This work led to numerous publications ( [22], [21], =-=[20]-=-). Several licences have been sold to companies. we also did a technical presentation of the GigaVoxels tool during Afig conference [17] in order to invit the community to use the tool. Figure 4. Giga... |

20 | Metric Learning for Large Scale Image Classification: Generalizing to New Classes at NearZero Cost
- Mensink, Verbeek, et al.
- 2012
(Show Context)
Citation Context .... Metric Learning for Large Scale Image Classification: Generalizing to new classes at near-zero cost Participants: Gabriela Csurka [XRCE], Thomas Mensink, Florent Perronnin [XRCE], Jakob Verbeek. In =-=[18]-=-, [27] we consider the task of large scale image classification in open ended datasets. Many real-life datasets are open-ended and dynamic: new images are continuously added to existing classes, new c... |

19 | Large-scale image classification with trace-norm regularization
- Harchaoui, Douze, et al.
- 2012
(Show Context)
Citation Context ... on sparse or dense local SIFT features. The dense feature extraction was optimized, so that they can be computed in real time on video data. The implementation was used for several publications [6], =-=[16]-=- and in our submisssion to the Trecvid 2012 MED task [31]. We provide a binary version of the local descriptor implementation, and the Fisher implementation is integrated in the Yael library, with Pyt... |

18 | Recognizing activities with cluster-trees of tracklets
- Gaidon, Harchaoui, et al.
(Show Context)
Citation Context ... improves the performance of a detector trained from still images alone. 6.3.3. Recognizing activities with cluster-trees of tracklets Participants: Adrien Gaidon, Zaid Harchaoui, Cordelia Schmid. In =-=[15]-=- we address the problem of recognizing complex activities, such as pole vaulting, which are characterized by the composition of a large and variable number of different spatio-temporal parts. We repre... |

17 | A survey of nonlinear prefiltering methods for efficient and accurate surface shading - Bruneton, Neyret - 2012 |

16 |
Explicit modeling of human-object interactions in realistic videos
- Prest, Ferrari, et al.
(Show Context)
Citation Context ... appointed as assistant professor at University of Edinburgh. V. Ferrari and C. Schmid co-supervised a PhD student (A. Prest) on the subject of automatic learning of objects in images and videos [3], =-=[9]-=-, [20]. A. Prest was bi-localized between ETH Zürich and Inria Grenoble. 8.3.3. Participation In International Programs • France-Berkeley fund: The LEAR team was awarded a grant from the France-Berkel... |

16 | Estimation procedures for semiparametric family of bivariate copulas
- Amblard, Girard
- 2005
(Show Context)
Citation Context ...ts: Stéphane Girard, Jonathan El-Methni, El-Hadji Deme. Joint work with: Guillou, A. and Gardes, L. (Univ. Strasbourg). We introduced a new model of tail distributions depending on two parameters τ ∈ =-=[0, 1]-=- and θ > 0. This model includes very different distribution tail behaviors from Fréchet and Gumbel maximum domains of attraction. In the particular cases of Pareto type tails (τ = 1) or Weibull tails ... |

16 | Extreme values and Haar series estimates of point processes boundaries
- Girard, Jacob
- 2003
(Show Context)
Citation Context ...s then an extreme quantile curve estimation problem. We proposed estimators based on projection as well as on kernel regression methods applied on the extreme values set, for particular set of points =-=[10]-=-. In collaboration with A. Daouia, we investigate the application of such methods in econometrics [42], [48]: A new characterization of partial boundaries of a free disposal multivariate support is in... |

14 | HDclassif: an R package for model-based clustering and discriminant analysis of high-dimensional data
- Bergé, Bouvreyon, et al.
- 2012
(Show Context)
Citation Context ...oach has been adapted to the unsupervised classification framework, and the related method is named High Dimensional Data Clustering (HDDC) [3]. Also, the description of the R package is published in =-=[11]-=-. Our recent work consists in adding a kernel in the previous methods to deal with nonlinear data classification [27], [45]. 6.1.2. Robust mixture modelling using skewed multivariate distributions wit... |

12 |
Keypoints and Local Descriptors of Scalar Functions on 2D Manifolds
- Zaharescu, Boyer, et al.
- 2012
(Show Context)
Citation Context ...rimental evaluation. The repeatability of the feature detector and the robustness of feature descriptor are tested, by applying a large number of deformations to the manifold or to the scalar function=-=[4]-=-. 7. Bilateral Contracts and Grants with Industry 7.1. Contracts with Technicolor A three year collaboration with Technicolor has started in 2011. The objective of this collaboration is to consider th... |

12 | Local higher-order statistics (lhs) for texture categorization and facial analysis
- Sharma, Hussain, et al.
- 2012
(Show Context)
Citation Context ...4.5 [21] S. Maji and A. Berg. Max-margin additive classifiers for detection. In ICCV,2009. 2 [22] M. Marszalek and C. Schmid. Constructing category hierarchies for visual recognition. In ECCV,2008. 2 =-=[23]-=- S. Nowozin and C. Lampert. Structured Learning and Prediction in Computer Vision. FoundationsandTrendsinComputerGraphicsand Vision, 2011. 1 [24] F. Perronnin and C. Dance. Fisher kernels on visual vo... |

12 | HOLZSCHUCH N.: Accurate fitting of measured reflectances using a shifted gamma micro-facet distribution - BAGHER, SOLER |

10 | Correlation-based burstiness for logo retrieval
- Revaud, Douze, et al.
- 2012
(Show Context)
Citation Context ...Retrieval Participants: Matthijs Douze, Jerome Revaud, Cordelia Schmid. Detecting logos in photos is challenging. A reason is that logos locally resemble patterns frequently seen in random images. In =-=[21]-=- we propose to learn a statistical model for the distribution of incorrect detections output by an image matching algorithm. It results in a novel scoring criterion in which the weight of correlated k... |

10 | Combining Monte Carlo and mean-field like methods for inference in hidden Markov random fields - Forbes, Fort - 2007 |

9 | NEYRET F.: Representing Appearance and Pre-Filtering Subpixel Data in Sparse Voxel Octrees
- HEITZ
(Show Context)
Citation Context ...epth of field to improve depth perception in direct volumetric rendering. His work has been accepted for publication at IEEE Pacific Visualization [15]. 2.3. Highlights of the Year BEST PAPER AWARD : =-=[16]-=- Representing Appearance and Pre-filtering Subpixel Data in Sparse Voxel Octrees in High Performance Graphics. E. HEITZ, F. NEYRET. 3. Scientific Foundations 3.1. Introduction The Maverick project-tea... |

8 |
Accurate object recognition with shape masks
- Marszalek, Schmid
(Show Context)
Citation Context .... ILSVRC 2010. http://www.imagenet.org/challenges/LSVRC/2010/index. 1, 5 [6] A. Bordes, L. Bottou, P. Gallinari, and J. Weston. Solving multiclass support vector machines with LaRank. In ICML, 2007.1 =-=[7]-=- L. Bottou. SGD. http://leon.bottou.org/projects/sgd. 4, 6 [8] L. Bottou and O. Bousquet. The tradeoffs of large scale learning. In NIPS,2007. 1, 4 [9] K. Chatfield, V. Lempitsky, A. Vedaldi, and A. Z... |

8 | Interactive rendering of acquired materials on dynamic geometry using bandwidth prediction - BAGHER, SOLER, et al. - 2012 |

8 | On kernel smoothing for extremal quantile regression
- Daouia, Gardes, et al.
- 2013
(Show Context)
Citation Context ...quantiles. When the covariate is functional and random (random design) and the tail of the distribution is heavy, we focus on kernel methods [18]. We extension to all kind of tails in investigated in =-=[15]-=-. Conditional extremes are studied in climatology where one is interested in how climate change over years might affect extreme temperatures or rainfalls. In this case, the covariate is univariate (ti... |

7 | Conditional gradient algorithms for machine learning
- Harchaoui, Juditsky, et al.
- 2012
(Show Context)
Citation Context ...ontent in the masked areas.Project-Team LEAR 11 6.2.2. Conditional gradient algorithms for machine learning Participants: Zaid Harchaoui, Anatoli Juditsky [UJF], Arkadi Nemirovski [Georgia Tech]. In =-=[17]-=- we consider convex optimization problems arising in machine learning in high-dimensional settings. For several important learning problems, such as e.g. noisy matrix completion, state-of-the-art opti... |

7 | Triplet Markov fields for the supervised classification of complex structure data - BLANCHET, FORBES |

7 |
High-dimensional data clustering, Computational Statistics and Data Analysis 52
- Bouveyron, Girard, et al.
- 2007
(Show Context)
Citation Context ...he supervised classification of objects in images. This approach has been adapted to the unsupervised classification framework, and the related method is named High Dimensional Data Clustering (HDDC) =-=[3]-=-. Also, the description of the R package is published in [11]. Our recent work consists in adding a kernel in the previous methods to deal with nonlinear data classification [27], [45]. 6.1.2. Robust ... |

7 |
Fast joint detection-estimation of evoked brain activity in event-related fMRI using a variational approach
- Chaari, Vincent, et al.
- 2013
(Show Context)
Citation Context ...M-JDE is robust to model mis-specification and provides computational gain while maintaining good performance in terms of activation detection and hemodynamic shape recovery. Main corresponding paper =-=[13]-=- 6.2.2. Hemodynamic-informed parcellation of fMRI data in a Joint Detection Estimation framework Participants: Florence Forbes, Lotfi Chaari, Thomas Vincent. Joint work with: Philippe Ciuciu from Team... |

6 | Tree-structured CRF models for interactive image labeling
- Mensink, Verbeek, et al.
(Show Context)
Citation Context ... we optimize the saliency keeping w fixed. 6.1.4. Tree-structured CRF Models for Interactive Image Labeling Participants: Gabriela Csurka [XRCE], Thomas Mensink, Jakob Verbeek.Project-Team LEAR 9 In =-=[8]-=- we propose structured prediction models for image labeling that explicitly take into account dependencies among image labels. In our tree structured models, image labels are nodes, and edges encode d... |

6 | Contextual object detection using setbased classification
- Cinbis, Sclaroff
- 2012
(Show Context)
Citation Context ... the 130K- [11] J. Deng, A. Berg, K. Li, and L. Fei-Fei. What does classifying more dim FV on ILSVRC 2010, it has a significant impact on than 10,000 image categories tell us? In ECCV,2010.1, 2, 5, 7 =-=[12]-=- J. Deng, W. Dong, R. Socher, L.-J. Li, K. Li, and L. Fei-Fei. ImaImageNet10K. This is not in contradiction with our pregenet: A large-scale hierarchical image database. In CVPR,2009.1, vious statemen... |

6 | Active Strokes: Coherent Line Stylization for Animated 3D Models - BÉNARD, JINGWAN, et al. - 2012 |

6 |
Functional kernel estimators of large conditional quantiles
- Gardes, Girard
- 2012
(Show Context)
Citation Context ...tors of the conditional tail index and conditional extreme quantiles. When the covariate is functional and random (random design) and the tail of the distribution is heavy, we focus on kernel methods =-=[18]-=-. We extension to all kind of tails in investigated in [15]. Conditional extremes are studied in climatology where one is interested in how climate change over years might affect extreme temperatures ... |

6 | Adaptive hemodynamic-informed parcellation of fMRI data in a variational joint detection estimation framework
- Chaari, Forbes, et al.
- 2012
(Show Context)
Citation Context ...ions. Validation on synthetic and real fMRI data demonstrate the JPDE performance over standard detection estimation schemes and suggest it as a new brain exploration tool. Corresponding papers [29], =-=[28]-=-. 6.2.3. Variational variable selection to assess experimental condition relevance in event-related fMRI Participants: Florence Forbes, Christine Bakhous, Lotfi Chaari, Thomas Vincent, Farida Enikeeva... |

5 |
Progressive shape models
- LETOUZEY, BOYER
- 2012
(Show Context)
Citation Context ...f a dynamic scene evolutions. The approach has been successfully experimented on several standard 4D datasets and we believe that it paves the way to more general multi-view scene capture and analysis=-=[8]-=-. Figure 4. Progressive Shape Models : the balloon can be separated from humans 6.4. Principal Geodesic Dynamics This work presents a new integration of a data-driven approach using dimension reductio... |

5 |
Frontier estimation with kernel regression on high order moments
- Girard, Guillou, et al.
- 2013
(Show Context)
Citation Context ...ndary are introduced. The regression is performed on the whole set of points, the selection of the “highest” points being automatically performed by the introduction of high order moments [19], [20], =-=[21]-=-. 6.4.7. Quantifying uncertainties on extreme rainfall estimations Participant: Stéphane Girard. Joint work with: Carreau, J. (Hydrosciences Montpellier), Gardes, L. (univ. Strasbourg) and Molinié, G.... |

5 | Kernel discriminant analysis and clustering with parsimonious Gaussian process models
- Bouveyron, Fauvel, et al.
- 2014
(Show Context)
Citation Context ...Data Clustering (HDDC) [3]. Also, the description of the R package is published in [11]. Our recent work consists in adding a kernel in the previous methods to deal with nonlinear data classification =-=[27]-=-, [45]. 6.1.2. Robust mixture modelling using skewed multivariate distributions with variable amounts of tailweight Participants: Florence Forbes, Darren Wraith. Clustering concerns the assignment of ... |

4 |
P.: N-tuple color segmentation for multi-view silhouette extraction
- Djelouah, Franco, et al.
- 2012
(Show Context)
Citation Context ...egmented based on their attached color models. The approach is significantly simpler and faster than previous multi-view segmentation methods, while providing results of equivalent or better quality. =-=[6]-=-. 6.7. Cage-based Motion Recovery using Manifold Learning10 Activity Report INRIA 2012 We have proposed a flexible model-based approach for the recovery of parameterized motion from a sequence of 3D ... |

4 | Real-time rendering of rough refraction - Rousiers, Bousseau, et al. - 2012 |

3 |
Estimating an endpoint with high order moments
- Girard, Guillou, et al.
- 2012
(Show Context)
Citation Context ...s of the boundary are introduced. The regression is performed on the whole set of points, the selection of the “highest” points being automatically performed by the introduction of high order moments =-=[19]-=-, [20], [21]. 6.4.7. Quantifying uncertainties on extreme rainfall estimations Participant: Stéphane Girard. Joint work with: Carreau, J. (Hydrosciences Montpellier), Gardes, L. (univ. Strasbourg) and... |

2 |
Segmentation of temporal mesh sequences into rigidly moving components. Graphical Models 75
- Arcila, Cagniart, et al.
- 2012
(Show Context)
Citation Context ...ovide tools for the evaluation of the approach, we also introduce new criteria to quantify a mesh segmentation. Results on both synthetic and real data as well as comparisons are provided in the paper=-=[3]-=-. 6.9. Keypoints and Local Descriptors of Scalar Functions on 2D Manifolds This work addresses the problem of describing surfaces using local features and descriptors. While methods for the detection ... |

2 |
Cage-based Motion Recovery using
- Duveau, Courtemanche, et al.
- 2012
(Show Context)
Citation Context ...y visual hull. Motion recovery applications are illustrated with a motion transfer encoding and the extraction of trajectories of anatomical joints. Validation is performed on the HumanEva II database=-=[7]-=-. Figure 6. Cage-based Motion Recovery using Manifold Learning 6.8. Segmentation of temporal mesh sequences into rigidly moving components This work considers the segmentation of meshes into rigid com... |

2 | A minimal solution for camera calibration using independent pairwise correspondences
- Vasconcelos, Barreto, et al.
- 2012
(Show Context)
Citation Context .... This makes the algorithm, for the specified use cases, probably the most suited calibration solution that does not require a calibration target, and hence can be performed without human interaction =-=[10]-=-. 6.6. N-Tuple Color Segmentation for Multi-View Silhouette Extraction We have presented a new method to extract multiple segmentations of an object viewed by multiple cameras, given only the camera c... |

2 | A discrete 3D+t Laplacian framework for mesh animation processing
- Hétroy
- 2012
(Show Context)
Citation Context ...e discrete 3D Laplacian, as soon as this parameter is non negative. We demonstrate the usefulness of this framework by extending Laplacian-based mesh editing and fairing techniques to mesh animations =-=[15]-=-. Figure 3. 3D+t Laplacian 6.2. Surface Flow Recovering dense motion information is a fundamental intermediate step in the image processing chain upon which higher level applications can be built, suc... |

2 | Large scale metric learning for distance-based image classification
- Mensink, Verbeek, et al.
(Show Context)
Citation Context ...ic Learning for Large Scale Image Classification: Generalizing to new classes at near-zero cost Participants: Gabriela Csurka [XRCE], Thomas Mensink, Florent Perronnin [XRCE], Jakob Verbeek. In [18], =-=[27]-=- we consider the task of large scale image classification in open ended datasets. Many real-life datasets are open-ended and dynamic: new images are continuously added to existing classes, new classes... |

2 | Learning latent factor models of human travel
- Guerzhoy, Hertzmann
(Show Context)
Citation Context ...nts: Gabriela Csurka [XRCE], Thomas Mensink, Jakob Verbeek. [29] J. Sánchez and F. Perronnin. High-dimensional signature compression for large-scale image classification. In CVPR, 2011. 1, 2, 4, 5, 7 =-=[30]-=- S. Shalev-Shwartz, Y. Singer, and N. Srebro. Pegasos: Primal estimate sub-gradient solver for SVM. In ICML, 2007.1 [31] A. Tewari and P. L. Bartlett. On the Consistency of Multiclass Classification M... |

2 | Real-time realistic rendering and lighting of forests - BRUNETON, NEYRET - 2012 |

2 |
Evaluation of classical spatial-analysis schemes of extreme rainfall, Nat. Hazards Earth Syst
- Ceresetti, Ursu, et al.
- 2012
(Show Context)
Citation Context ...is region that can be obtained. It is then of primary importance to study the sensitivity of the extreme rainfall estimation to the estimation method considered. The obtained results are published in =-=[12]-=-. 6.4.8. Retrieval of Mars surface physical properties from OMEGA hyperspectral images. Participant: Stéphane Girard. Joint work with: Douté, S. from Laboratoire de Planétologie de Grenoble, France an... |

2 | Characteristics of Acacia mangium shoot apical meristems in natural and in vitro conditions in relation to heteroblasty, in "Trees - Structure and Function - HATT, MANKESSI, et al. |

2 |
Adaptive experimental condition selection in event-related fMRI
- BAKHOUS, FORBES, et al.
- 2012
(Show Context)
Citation Context ...tion detection and local HRF estimation that we illustrate on synthetic and real fMRI data. This approach is an alternative to our previous approach based on Monte-Carlo Markov Chain (MCMC) inference =-=[25]-=-. Corresponding paper [26]. 6.2.4. Bayesian BOLD and perfusion source separation and deconvolution from functional ASL imaging Participants: Florence Forbes, Thomas Vincent. In the context of ARC AINS... |

2 | An improved cuda-based implementation of differential evolution on gpu
- Qin, Raimondo, et al.
- 2012
(Show Context)
Citation Context ...s. This can be interesting in image processing for instance where it can be difficult to introduce parametric models that are general enough (e.g. for contours). 2.2. Highlights of the Year Our paper =-=[33]-=- entitled An Improved CUDA-Based Implementation of Differential Evolution on GPU was nominated and finalist for the best paper award in the Digital Entertainment Technologies and Arts / Parallel Evolu... |

1 |
Acquisition de surfaces déformables à partir d’un système multicamera calibré, Université de Grenoble
- CAGNIART
- 2012
(Show Context)
Citation Context ... 12h, Ensimag 3rd year, Grenoble INP PhD: Cédric Cagniart, Motion Capture of Deformable Surfaces in Multi-View Studios, Université de Grenoble, supervised by Edmond Boyer and Slobodan Ilic, 16/07/12, =-=[1]-=-. PhD: Estelle Duveau, Motion Measurement of Small Vertebrates, Université de Grenoble, supervised by Lionel Reveret and Edmond Boyer, 03/12/12, non publically released, patent pending. PhD: Antoine L... |

1 |
Modélisation 4D à partir de plusieurs caméras, Université de Grenoble
- LETOUZEY
(Show Context)
Citation Context ...nd Edmond Boyer, 03/12/12, non publically released, patent pending. PhD: Antoine Letouzey, Modélisation 4D à partir de plusieurs caméras, Universitè de Grenoble, supervised by Edmond Boyer, 30/07/12, =-=[2]-=-. PhD in progress : Benjamin Allain, Geometry and Appearance Analysis of Deformable 3D shapes, Université de Grenoble, started 01/10/2012, supervised by J.-S. Franco and E. Boyer. PhD in progress : Be... |

1 | Flot de scène, in "Traitement du Signal", 2013, Ce travail a été partiellement financé par OSEO, l’agence française pour l’innovation, dans le cadre du programme de recherche QUAERO. A paraître en 2013 - LETOUZEY, PETIT, et al. |

1 | Modèles progressifs de forme, in "CORESA - COmpression et REprésentation des Signaux Audiovisuels - 2012 - LETOUZEY, BOYER - 2012 |

1 |
Flot de scène à partir d’images couleur et de cartes de profondeur, in "RFIA 2012 (Reconnaissance des Formes et Intelligence Artificielle
- LETOUZEY, PETIT, et al.
(Show Context)
Citation Context ...ch photometric constraints can be consistently integrated in 3D. Experiments on real and synthetic data provide both qualitative and quantitative results that demonstrated the interest of the approach=-=[12]-=-. 6.3. Progressive Shape Models In this work we address the problem of recovering both the topology and the geometry of a deformable shape using temporal mesh sequences (Figure 4). The interest arises... |

1 | Sensorimotor characteristics of speech production, in "Audiovisual Speech Processing - BAILLY, BADIN, et al. |

1 | Temporally coherent mesh sequence segmentations, Inria, January 2012, n o RR-7856 - ARCILA, CAGNIART, et al. |

1 | Structured Models for Action Recognition in Real-world Videos - Modèles Structurés pour la Reconnaissance d’Actions dans des Vidéos Réalistes, Université de Grenoble
- GAIDON
- 2012
(Show Context)
Citation Context ...theses defended in 2012: • • • • 9.2.3. Juries A. Gaidon, Structured Models for Action Recognition in Real-world Videos - Modèles Structurés pour la Reconnaissance d’Actions dans des Vidéos Réalistes =-=[1]-=-, Université de Grenoble, 25/10/2012, advisers: Z. Harchaoui and C. Schmid. T. Mensink, Apprentissage de Modèles pour la Classification et la Recherche d’Images [2], Université de Grenoble, 26/10/2012... |

1 |
Apprentissage de Modèles pour la Classification et la Recherche d’Images, Université de Grenoble
- MENSINK
- 2012
(Show Context)
Citation Context ...d’Actions dans des Vidéos Réalistes [1], Université de Grenoble, 25/10/2012, advisers: Z. Harchaoui and C. Schmid. T. Mensink, Apprentissage de Modèles pour la Classification et la Recherche d’Images =-=[2]-=-, Université de Grenoble, 26/10/2012, advisers: J. Verbeek, G. Csurka, and C. Schmid. A. Prest, Weakly supervised methods for learning actions and objects [3], ETHZ, 4/9/2012, advisers: V. Ferrari, an... |

1 | Weakly supervised methods for learning actions and objects, Eidgenössische Technische Hochschule Zürich (ETHZ
- PREST
- 2012
(Show Context)
Citation Context ...ently appointed as assistant professor at University of Edinburgh. V. Ferrari and C. Schmid co-supervised a PhD student (A. Prest) on the subject of automatic learning of objects in images and videos =-=[3]-=-, [9], [20]. A. Prest was bi-localized between ETH Zürich and Inria Grenoble. 8.3.3. Participation In International Programs • France-Berkeley fund: The LEAR team was awarded a grant from the France-B... |

1 |
Semantic description of humans in images, Université de Caen
- SHARMA
(Show Context)
Citation Context ...obert, O. Chapelle, and K. Weinberger. Supervised semantic indexing. In CIKM, 2009.4 [3] P. L. Bartlett, M. I. Jordan, and J. D. McAuliffe. Convexity, classification, and risk bounds. In NIPS, 2003.7 =-=[4]-=- S. Bengio, J. Weston, and D. Grangier. Label embedding trees for large multi-class tasks. In NIPS, 2010.2 [5] A. Berg, J. Deng, and L. Fei-Fei. ILSVRC 2010. http://www.imagenet.org/challenges/LSVRC/2... |

1 | Fusion of speech, faces and text for person identification
- BREDIN, POIGNANT, et al.
- 2012
(Show Context)
Citation Context ...w-OVR and 130K-dim Fisher vectors. (a-d) Sample classes among the best performing ones. (e-h) Sample classes among the worst performing ones. while rebalancing the data had little impact on the 130K- =-=[11]-=- J. Deng, A. Berg, K. Li, and L. Fei-Fei. What does classifying more dim FV on ILSVRC 2010, it has a significant impact on than 10,000 image categories tell us? In ECCV,2010.1, 2, 5, 7 [12] J. Deng, W... |

1 |
Tree-walk kernels for computer vision, in "Image Processing and Analysis with Graphs: Theory and Practice
- HARCHAOUI, BACH
(Show Context)
Citation Context ...ificantly better than state-of-the-art approaches for Fisher vectors with 16 Gaussians. 6.2.4. Tree-walk kernels for computer vision Participants: Francis Bach [Inria SIERRA team], Zaid Harchaoui. In =-=[25]-=- we propose a family of positive-definite kernels between images, allowing to compute image similarity measures respectively in terms of color and of shape. The kernels consists in matching subtree-pa... |

1 |
Action Detection with Actom Sequence Models
- GAIDON, HARCHAOUI, et al.
(Show Context)
Citation Context ...m Sequence Models Participants: Adrien Gaidon, Zaid Harchaoui, Cordelia Schmid. We address the problem of detecting actions, such as drinking or opening a door, in hours of challenging video data. In =-=[26]-=- we propose a model based on a sequence of atomic action units, termed "actoms", that are semantically meaningful and characteristic for the action. Our Actom Sequence Model (ASM) represents the tempo... |

1 |
Kernel change-point detection, 2012, Available online at arXiv
- ARLOT, CELISSE, et al.
(Show Context)
Citation Context ...d in the other step we optimize the saliency keeping w fixed. 809 6.1.4. Tree-structured CRF Models for Interactive Image Labeling Participants: Gabriela Csurka [XRCE], Thomas Mensink, Jakob Verbeek. =-=[29]-=- J. Sánchez and F. Perronnin. High-dimensional signature compression for large-scale image classification. In CVPR, 2011. 1, 2, 4, 5, 7 [30] S. Shalev-Shwartz, Y. Singer, and N. Srebro. Pegasos: Prima... |

1 | Material appearance: photorealistic representation and rendering, Université de Grenoble - BAGHER - 2012 |

1 | A Frequency Analysis of Light Transport: from Theory to Implementation, Université de Grenoble, November 2012. Activity Report INRIA - BELCOUR |

1 | Visualisation interactive de grands volumes de données incertaines : pour une approche perceptive, Université de Grenoble - CONINX - 2012 |

1 | Multi-scale Feature-Preserving Smoothing of Images and - JIBAI |

1 | Immersive Virtual Environment for VisuoVestibular Therapy: Preliminary Results - GASCUEL, PAYNO, et al. |

1 | Volume Preserving FFD for Programmable Graphics Hardware, in "Visual Computer
- HAHMANN, BONNEAU, et al.
- 2012
(Show Context)
Citation Context ...oration with Gershon Elber from TECHNION, Hans Hagen from TU Kaiserslautern, Georges-Pierre Bonneau and Sébastien Barbier from Maverick Inria. It has been published in the journal The Visual Computer =-=[5]-=-. Figure 15. Comparison between standard FFD deformation (middle) and our method preserving the volume (right) from an initial rest shape (left). Within Lucian Stanculescu PhD, we developped a mesh st... |

1 | feature preserving MLS surface reconstruction based on local feature line approximations, in "Graphical Models - Sharp - 2012 |

1 |
Evaluation of Depth of Field for Depth Perception
- GROSSET, SCHOTT, et al.
- 2013
(Show Context)
Citation Context ...tric experiment in order to evaluate the benefit of depth of field to improve depth perception in direct volumetric rendering. His work has been accepted for publication at IEEE Pacific Visualization =-=[15]-=-. 2.3. Highlights of the Year BEST PAPER AWARD : [16] Representing Appearance and Pre-filtering Subpixel Data in Sparse Voxel Octrees in High Performance Graphics. E. HEITZ, F. NEYRET. 3. Scientific F... |

1 |
GigaVoxels, librairie et kit de développement sur GPU pour l’exploration temps-réel et visuellement réaliste d’immenses scènes détaillées à base de SVO, in "Journées de l’Association Française d’Informatique Graphique
- CRASSIN, GUEHL, et al.
(Show Context)
Citation Context ...plex data. This work led to numerous publications ( [22], [21], [20]). Several licences have been sold to companies. we also did a technical presentation of the GigaVoxels tool during Afig conference =-=[17]-=- in order to invit the community to use the tool. Figure 4. GigaVoxels freely downloadable demo. 6. New Results 6.1. Computer visualization 6.1.1. Immersive Virtual Environment for Visuo-Vestibular Th... |

1 | Open Archives and the Invisible College: driving forces towards a new publication model, in "Dagstuhl Perspectives Workshop: Publication Culture in Computing Research - HOLZSCHUCH |

1 | Temporally Coherent Video Stylization, in "Image and Videobased Artistic Stylisation - BÉNARD, THOLLOT, et al. - 2012 |

1 |
On the difficulty to clearly identify and delineate disease risk hot spots, in "International Journal of Applied Earth Observation and Geoinformation", May 2012, available on line
- CHARRAS-GARRIDO, AZIZI, et al.
(Show Context)
Citation Context ...how that our strategy performs well in dealing with low population sizes and accurately determines high risk regions, both in terms of localization and risk level estimation. Main corresponding paper =-=[14]-=-. 6.3.2. Spatial modelling of biodiversity from high-througput DNA sequence data Participants: Florence Forbes, Angelika Studeny. This is joint work with Eric Coissac and Pierre Taberlet from LECA (La... |

1 |
Optimization of power consumption and user impact based on point process modeling of the request sequence
- DURAND, GIRARD, et al.
- 2011
(Show Context)
Citation Context ...aborations with the EDF research department and the LCFR laboratory (Laboratoire de Conduite et Fiabilité des Réacteurs) of CEA / Cadarache. We also consider failure detection in print infrastructure =-=[16]-=- through collaboration with Xerox, Meylan. 5. Software 5.1. The ECMPR software Participant: Florence Forbes. Joint work with: Radu Horaud and Manuel Iguel. The ECMPR (Expectation Conditional Maximizat... |

1 |
Estimation of extreme quantiles from heavy and light tailed distributions
- EL-METHNI, GARDES, et al.
- 2011
(Show Context)
Citation Context ... work of Jonathan El-methni has been dedicated to the definition of an estimator of the parameter τ. This permits the construction of new estimators of extreme quantiles. The results are published in =-=[17]-=-. Our future work will consist in proposing a test procedure in order to discriminate between Pareto and Weibull tails. We are also working on the estimation of the second order parameter ρ (see parag... |

1 |
Statistical estimation of dominant physical parameters for leakage variability in 32nanometer CMOS under supply voltage variations, in "Journal of Low Power Electronics
- JOSHI, LOMBARDOT, et al.
(Show Context)
Citation Context ... industrial design flow. Our objective to model the total consumption of the circuit from the probability distribution of consumption of each individual gate. Our preliminary results are published in =-=[23]-=-. 7. Partnerships and Cooperations 7.1. Regional Initiatives MISTIS participates in the weekly statistical seminar of Grenoble. F. Forbes is one of the organizers and several lecturers have been invit... |

1 | Algebraic properties of copulas defined from matrices - AMBLARD, GIRARD, et al. - 2012 |

1 | Variational Variable Selection To Assess Experimental Condition Relevance
- BAKHOUS, FORBES, et al.
- 2013
(Show Context)
Citation Context ...RF estimation that we illustrate on synthetic and real fMRI data. This approach is an alternative to our previous approach based on Monte-Carlo Markov Chain (MCMC) inference [25]. Corresponding paper =-=[26]-=-. 6.2.4. Bayesian BOLD and perfusion source separation and deconvolution from functional ASL imaging Participants: Florence Forbes, Thomas Vincent. In the context of ARC AINSI project, joint work with... |

1 | Robust voxel-wise Joint Detection Estimation of brain activity in fMRI
- CHAARI, FORBES, et al.
(Show Context)
Citation Context ...oximations. Validation on synthetic and real fMRI data demonstrate the JPDE performance over standard detection estimation schemes and suggest it as a new brain exploration tool. Corresponding papers =-=[29]-=-, [28]. 6.2.3. Variational variable selection to assess experimental condition relevance in event-related fMRI Participants: Florence Forbes, Christine Bakhous, Lotfi Chaari, Thomas Vincent, Farida En... |

1 | Statistical leakage estimation in 32nm CMOS considering cells correlations - JOSHI, LOMBARDOT, et al. - 2012 |

1 |
Augmented cumulative distribution networks for multivariate extreme value modelling
- MAZO, FORBES, et al.
- 2012
(Show Context)
Citation Context ... max-stable CDN’s. To face this limitation, we therefore propose to augment max-stable CDN’s with the more standard Gumbel max-stable distribution function in order to enrich the dependence structure =-=[32]-=-. 6.4.4. Modelling extremal events Participants: Stéphane Girard, Jonathan El-Methni, El-Hadji Deme. Joint work with: Guillou, A. and Gardes, L. (Univ. Strasbourg). We introduced a new model of tail d... |

1 | Sélection de variable dans un cadre bayésien de traitement de données d’IRM fonctionnelle, in "Journées de Statistique de la Société Française de Statistique (SFdS - BAKHOUS, FORBES, et al. |

1 | Processus gaussiens parcimonieux pour la classification générative de données hétérogènes, in "44èmes Journées de Statistique de la Société Française de Statistique - BOUVEYRON, FAUVEL, et al. - 2012 |

1 | Parcel-free Joint Detection-Estimation in fMRI, in "Journées de Statistique de la Société Française de Statistique (SFdS - CHAARI, FORBES, et al. |

1 | Estimation de l’espérance conditionnelle des pertes extrêmes dans le cas de lois à queues lourdes en présence d’une covariable, in "44èmes Journées de Statistique organisées par la Société Française de Statistique", Bruxelles, Belgique, mai 2012 - EL-METHNI, GARDES, et al. |

1 | Estimation de point terminal dans le domaine d’attraction de Weibull par une méthode des moments d’ordre élevé, in "44èmes Journées de Statistique organisées par la Société Française de Statistique", Bruxelles, Belgique, mai 2012 - GIRARD, GUILLOU, et al. |