Showing 1 to 21 of 21 matching Articles
Results per page:
Export (CSV)
By
Ramírez, Justino; Rivera, Mariano
We present an algorithm for automatic selection of features that best segment an image in texture homogeneous regions. The set of “best extractors” are automatically selected among the Gabor filters, Cooccurrence matrix, Law’s energies and intensity response. Noisefeatures elimination is performed by taking into account the magnitude and the granularity of each feature image, i.e. the compute image when a specific feature extractor is applied. Redundant features are merged by means of probabilistic rules that measure the similarity between a pair of image feature. Then, cascade applications of general purpose image segmentation algorithms (KMeans, GraphCut and ECGMMF) are used for computing the final segmented image. Additionally, we propose an evolutive gradient descent scheme for training the method parameters for a benchmark image set. We demonstrate by experimental comparisons, with stat of the art methods, a superior performance of our technique.
more …
By
Alba, Alfonso; ArceSantana, Edgar; Rivera, Mariano
3 Citations
Motion estimation is one of the most important tasks in computer vision. One popular technique for computing dense motion fields consists in defining a large enough set of candidate motion vectors, and assigning one of such vectors to each pixel, so that a given cost function is minimized. In this work we propose a novel method for finding a small set of adequate candidates, making the minimization process computationally more efficient. Based on this method, we present algorithms for the estimation of dense optical flow using two minimization approaches: one based on a classic blockmatching procedure, and another one based on entropycontrolled quadratic Markov measure fields which allow one to obtain smooth motion fields. Finally, we present the results obtained from the application of these algorithms to examples taken from the Middlebury database.
more …
By
Aranda, Ramón; Rivera, Mariano; RamírezManzanares, Alonso
We present a new regularization approach for Diffusion Basis Functions fitting to estimate in vivo brain the axonal orientation from Diffusion Weighted Magnetic Resonance Images. That method assumes that the observed Magnetic Resonance signal at each voxel is a linear combination of a given diffusion basis functions; the aim of the approach is the estimation of the coefficients of the linear combination. An issue with the Diffusion Basis Functions method is the overestimation on the number of tensors (associated with different axon fibers) within a voxel due to noise, namely, the over fitting of the noisy signal. Our proposal overcomes such an overestimation problem. In additionally, we propose a metric to compare the performance of multifiber estimation algorithms. The metric is based on the Earth Mover’s Distance and allows us to compare in a single metric the orientation, size compartment and the number of axon bundles between two different estimations. The improvements of our two proposals is shown on synthetic and real experiments.
more …
By
Madrigal, Francisco; Rivera, Mariano; Hayet, JeanBernard
5 Citations
This paper describes an original strategy for using a datadriven probabilistic motion model into particle filterbased target tracking on video streams. Such a model is based on the local motion observed by the camera during a learning phase. Given that the initial, empirical distribution may be incomplete and noisy, we regularize it in a second phase. The hybrid discretecontinuous probabilistic motion model learned this way is then used as a sampling distribution in a particle filter framework for target tracking. We present promising results for this approach in some common datasets used as benchmarks for visual surveillance tracking algorithms.
more …
By
RamírezManzanares, Alonso; Rivera, Mariano; Kornprobst, Pierre; Lauze, François
Show all (4)
4 Citations
We propose a variational approach for multivalued velocity field estimation in transparent sequences. Starting from existing local motion estimators, we show a variational model for integrating in space and time these local estimations to obtain a robust estimation of the multivalued velocity field. With this approach, we can indeed estimate some multivalued velocity fields which are not necessarily piecewise constant on a layer: Each layer can evolve according to nonparametric optical flow. We show how our approach outperforms some existing approaches, and we illustrate its capabilities on several challenging synthetic/real sequences.
more …
By
Dalmau, Oscar; Rivera, Mariano
1 Citations
We propose a new model for probabilistic image segmentation with spatial coherence through a Markov Random Field prior. Our model is based on a generalized information measure between discrete probability distribution (βMeasure). This model generalizes the quadratic Markov measure field models (QMMF). In our proposal, the entropy control is achieved trough the likelihood energy. This entropy control mechanism makes appropriate our method for being used in tasks that require of the simultaneous estimation of the segmentation and the model parameters.
more …
By
Dalmau, Oscar; Rivera, Mariano
3 Citations
We propose a general Bayesian model for image segmentation with spatial coherence through a Markov Random Field prior. We also study variants of the model and their relationship. In this work we use the Matusita Distance, although our formulation admits other metricdivergences. Our main contributions in this work are the following. We propose a general MRFbased model for image segmentation. We study a model based on the Matusita Distance, whose solution is found directly in the discrete space with the advantage of working in a continuous space. We show experimentally that this model is competitive with other models of the state of the art. We propose a novel way to deal with nonlinearities (irrational) related with the Matusita Distance. Finally, we propose an optimization method that allows us to obtain a hard image segmentation almost in real time and also prove its convergence.
more …
By
Madrigal, Francisco; Rivera, Mariano; Hayet, JeanBernard
This paper presents a particle filterbased approach for multiple target tracking in video streams in single static cameras settings. We aim in particular to manage middense crowds situations, where, although tracking is possible, it is made complicated by the presence of frequent occlusions among targets and with scene clutter. Moreover, the appearance of targets is sometimes very similar, which makes standard trackers often switch their target identity. Our contribution is twofold: (1) we first propose an estimation scheme for motion priors in the camera field of view, that integrates sparse optical flow data and regularizes the corresponding discrete distribution fields on velocity directions and magnitudes; (2) we use these motion priors in a hybrid motion model for a particle filter tracker. Through several results on videosurveillance datasets, we show the pertinence of this approach.
more …
By
Rivera, Mariano; Ocegueda, Omar; Marroquin, Jose L.
5 Citations
We present a computationally efficient segmentation–restoration method, based on a probabilistic formulation, for the joint estimation of the label map (segmentation) and the parameters of the feature generator models (restoration). Our algorithm computes an estimation of the posterior marginal probability distributions of the label field based on a Gauss Markov Random Measure Field model. Our proposal introduces an explicit entropy control for the estimated posterior marginals, therefore it improves the parameter estimation step. If the model parameters are given, our algorithm computes the posterior marginals as the global minimizers of a quadratic, linearly constrained energy function; therefore, one can compute very efficiently the optimal (Maximizer of the Posterior Marginals or MPM) estimator for multi–class segmentation problems. Moreover, a good estimation of the posterior marginals allows one to compute estimators different from the MPM for restoration problems, denoising and optical flow computation. Experiments demonstrate better performance over other state of the art segmentation approaches.
more …
By
Calderon, Felix; Ruiz, Ubaldo; Rivera, Mariano
4 Citations
Fastest threedimensional (3D) surface reconstruction algorithms, from point clouds, require of the knowledge of the surface–normals. The accuracy, of state of the art methods, depends on the precision of estimated surface–normals. Surface–normals are estimated by assuming that the surface can be locally modelled by a plane as was proposed by Hoppe et. al [1]. Thus, current methods for estimating surface–normals are prone to introduce artifacts at the geometric edges or corners of the objects. In this paper an algorithm for Normal Estimation with Neighborhood Reorganization (NENR) is presented. Our proposal changes the characteristics of the neighborhood in places with corners or edges by assuming a locally plane piecewise surface. The results obtained by NENR improve the quality of the normal with respect to the state of the art algorithms. The new neighborhood computed by NENR, use only those points that belong to the same plane and they are the nearest neighbors. Experiments in synthetic and real data shown an improvement on the geometric edges of 3D reconstructed surfaces when our algorithm is used.
more …
By
Aranda, Ramón; Rivera, Mariano; RamírezManzanares, Alonso; Ashtari, Manzar; Gee, James C.
Show all (5)
1 Citations
We propose a method for estimating axonal fiber connectivity pathways (cerebral connectivity fibers) from MultiTensor Diffusion Magnetic Resonance Imaging (MTDMRI). Our method uses multiple local orientation information provided by MTDMRI for leading stochastic walks of particles. We perform stochastic walks on particles with mass which introduce inertia and gravitational forces that result in filtered trajectories. Afterwards, the fiber bunches are estimated with a clustering procedure based on terminal points that allows us to eliminate outliers. The method’s performance is evaluated on MTDMRI from realistic synthetic data, a diffusion phantom and demonstrated in real human brain data.
more …
By
RamirezManzanares, Alonso; Rivera, Mariano; Gee, James C.
Diffusion weighted magnetic resonance imaging is widely used in the study of the structure of the fiber pathways in brain white matter. In this work we present a new method for denoising intra–voxel axon fiber tracks. In order to improve local (voxelwise) estimations, we use the general–purpose segmentation method called Entropy–Controlled Quadratic Markov Measure Field Models. Our proposal is capable of spatially–regularize multiple axon fiber orientations (intravoxel orientations). In order to provide the best as possible local axon orientations to our spatial regularization procedure, we evaluate two optimization methods for fitting a Diffusion Basis Function model. We present qualitative results on real human Diffusion Weighted MRI data where the ground–truth is not available, and we quantitatively validate our results by synthetic experiments.
more …
By
Calderon, Felix; Rivera, Mariano
In this paper we present an image filter based on proximity, range information and Surface Normal information, in order to distinguish discontinuities created by planes in different orientations. Our main contribution is the estimation of a piecewise smooth Surface Normal, the discontinuity for the Surface Normal and their use for image restoration. There are many applications for Surface Normals (SN) in many research fields, because it is a local measure of the surface orientation. The Bilateral Filter measure differences in range in order to weight a window around a point, this condition is equivalent to see the image as horizontal planes, nevertheless the image do not have the same orientation in different places so surface orientation could help to up perform the Bilateral Filter results. We present a Trilateral Filter (TF) based on proximity, range and Surface Normal information. In this paper, we present a robust algorithm to compute the SN and a new kernel based on SN, which does not have Gaussian formulation. With our Trilateral Filter we up perform the results obtained by BF and we shown with some experiments in which the images filter by our TF looks sharper than the image filter by BF.
more …
By
RamirezManzanares, Alonso; PalafoxGonzalez, Abel; Rivera, Mariano
Motion estimation in sequences with transparencies is an important problem in robotics and medical imaging applications. In this work we propose two procedures to improve the transparent optical flow computation. We build from a variational approach for estimating multivalued velocity fields in transparent sequences. That method estimates multivalued velocity fields which are not necessarily piecewise constant on a layer –each layer can evolve according to a nonparametric optical flow. First we introduce a robust statistical spatial interaction weight which allows to segment the multimotion field. As result, our method is capable to recover the object’s shape and the velocity field for each object with high accuracy. Second, we develop a procedure to separate the component layers of rigid objects from a transparent sequence. Such a separation is possible because of the high accuracy of the object’s shape recovered from our transparent optical flow computation. Our proposal is robust to the presence of several objects in the same sequence as well as different velocities for the same object along the sequence. We show how our approach outperforms existing methods and we illustrate its capabilities on challenging sequences.
more …
By
Rivera, Mariano; Gee, James
2 Citations
The object of this paper is to present a formulation for the segmentation and restoration problem using flexible models with a robust regularized network (RRN). A twosteps iterative algorithm is presented. In the first step an approximation of the classification is computed by using a local minimization algorithm, and in the second step the parameters of the RRN are updated. The use of robust potentials is motivated by (a) classification errors that can result from the use of local minimizer algorithms in the implementation, and (b) the need to adapt the RN using local image gradient information to improve fidelity of the model to the data.
more …
By
Rivera, Mariano; Marroquin, Jose L.
1 Citations
The propose of this paper is to introduce a new regularization formulation for inverse problems in computer vision and image processing that allows one to reconstruct second order piecewise smooth images, that is, images consisting of an assembly of regions with almost constant value, almost constant slope or almost constant curvature. This formulation is based on the idea of using potential functions that correspond to springs or thin plates with an adaptive rest condition. Efficient algorithms for computing the solution, and examples illustrating the performance of this scheme, compared with other known regularization schemes are presented as well.
more …
By
Madrigal, Francisco; Hayet, JeanBernard; Rivera, Mariano
2 Citations
This article describes an original strategy for enhancing current stateoftheart trackers through the use of motion priors, built as datadriven probabilistic motion models for moving targets. Our priors have a simple form and can replace advantageously more traditional models, such as the constant velocity or constant acceleration models, that are of common use in visual tracking systems, but that are also prone to fail in handling critical scenerelated constraints on the targets motion. These priors are learned based on local motion observed in the video stream(s) and, given that the obtained representation may be incomplete and noisy, we regularize it in a second phase. The hybrid discrete–continuous motion priors are then used within two classical target tracking approaches: (1) as a sampling distribution in a particle filter framework and (2) as a weighting prior in a detectionbased framework. For both tracking schemes, we present promising results with our motion prior approach, on classical benchmark datasets from the visual surveillance tracking literature.
more …
By
RamírezManzanares, Alonso; Rivera, Mariano; Kornprobst, Pierre; Lauze, François
Show all (4)
2 Citations
Motion estimation in sequences with transparencies is an important problem in robotics and medical imaging applications. In this work we propose a variational approach for estimating multivalued velocity fields in transparent sequences. Starting from existing local motion estimators, we derive a variational model for integrating in space and time such a local information in order to obtain a robust estimation of the multivalued velocity field. With this approach, we can indeed estimate multivalued velocity fields which are not necessarily piecewise constant on a layer—each layer can evolve according to a nonparametric optical flow. We show how our approach outperforms existing methods; and we illustrate its capabilities on challenging experiments on both synthetic and real sequences.
more …
By
HernandezLopez, Francisco J.; Rivera, Mariano
15 Citations
We present a method for foreground/background video segmentation (change detection) in realtime that can be used, in applications such as background subtraction or analysis of surveillance cameras. Our approach implements a probabilistic segmentation based on the Quadratic Markov Measure Field models. This framework regularizes the likelihood of each pixel belonging to each one of the classes (background or foreground). We propose a new likelihood that takes into account two cases: the first one is when the background is static and the foreground might be static or moving (Static Background Subtraction), the second one is when the background is unstable and the foreground is moving (Unstable Background Subtraction). Moreover, our likelihood is robust to illumination changes, cast shadows and camouflage situations. We implement a parallel version of our algorithm in CUDA using a NVIDIA Graphics Processing Unit in order to fulfill realtime execution requirements.
more …
By
Rivera, Mariano; Dalmau, Oscar; Mio, Washington
1 Citations
We present a framework for image segmentation based on the ML estimator. A common hypothesis for explaining the differences among image regions is that they are generated by sampling different Likelihood Functions. We adopt last hypothesis and, additionally, we assume that such samples are i.i.d. Thus, the probability of a model generates the observed pixel value is estimated by computing the likelihood of the sample composed with the surrounding pixels.
more …
By
RamirezManzanares, Alonso; Rivera, Mariano
19 Citations
We present a regularized method for solving an inverse problem in Diffusion Tensor Magnetic Resonance Imaging (DTMRI) data. In the case of brain images, DTMR imagery technique produces a tensor field that indicates the local orientation of nerve bundles. Now days, the spatial resolution of this technique is limited by the partial volume effect produced in voxels that contain fiber crossings or bifurcations. In this paper, we proposed a method for recovering the intravoxel information and inferring the brain connectivity. We assume that the observed tensor is a linear combination of a given tensor basis, therefore, the aim of our approach is the computation of the unknown coefficients of this linear combination. By regularizing the problem, we introduce the needed prior information about the piecewise smoothness of nerve bundles orientation. As a result, we recover a multitensor field. Moreover, for estimating the nerve bundles trajectory, we propose a method based on stochastic walks of particles through the computed multitensor field. The performance of the method is demonstrated by experiments in both synthetic and real data.
more …
