Results 1  10
of
17
Adaptive structure from motion with a contrario model estimation
 In Proc. ACCV
, 2012
"... Abstract. Structure from Motion (SfM) algorithms take as input multiview stereo images (along with internal calibration information) and yield a 3D point cloud and camera orientations/poses in a common 3D coordinate system. In the case of an incremental SfM pipeline, the process requires repeated mo ..."
Abstract

Cited by 10 (8 self)
 Add to MetaCart
(Show Context)
Abstract. Structure from Motion (SfM) algorithms take as input multiview stereo images (along with internal calibration information) and yield a 3D point cloud and camera orientations/poses in a common 3D coordinate system. In the case of an incremental SfM pipeline, the process requires repeated model estimations based on detected feature points: homography, fundamental and essential matrices, as well as camera poses. These estimations have a crucial impact on the quality of 3D reconstruction. We propose to improve these estimations using the a contrario methodology. While SfM pipelines usually have globally xed thresholds for model estimation, the a contrario principle adapts thresholds to the input data and for each model estimation. Our experiments show that adaptive thresholds reach a signi cantly better precision. Additionally, the user is free from having to guess thresholds or to optimistically rely on default values. There are also cases where a globally xed threshold policy, whatever the threshold value is, cannot provide the best accuracy, contrary to an adaptive threshold policy. 1
Local matching indicators for transport problems with concave costs
 SIAM J. Disc. Math
, 2012
"... ar ..."
(Show Context)
Wasserstein Active Contours
, 2011
"... Abstract. In this paper, we propose a novel and rigorous framework for regionbased active contours that combines the Wasserstein distance between statistical distributions in arbitrary dimension and shape derivative tools. To the best of our knowledge, this is the first variational image segmentati ..."
Abstract

Cited by 6 (0 self)
 Add to MetaCart
(Show Context)
Abstract. In this paper, we propose a novel and rigorous framework for regionbased active contours that combines the Wasserstein distance between statistical distributions in arbitrary dimension and shape derivative tools. To the best of our knowledge, this is the first variational image segmentation algorithm that involves regiondependent multidimensional descriptors based on the optimal transport theory. The distributions are represented owing to nonparametric kernel density estimators (e.g. Parzen), and the exact evolution speed corresponding to the Wassersteinbased segmentation energy is provided. To speedup the computation and be able to handle highdimensional features and largescale data, we introduce a sliced Wasserstein approximation of the original Wasserstein distance. The framework is flexible enough to allow either minimization of the Wasserstein distance to known fixed distributions, or maximization of the distance between the distributions of the regions to be segmented (region competition). Numerical results reported to show the advantages of the proposed optimal transport distance with respect to alternative metrics (such as the KullbackLeibler divergence). These traditional metrics cannot deal properly with distributions having localized supports, and do not take into account the distance between the modes of the histograms. Additionally, our framework handles distributions in arbitrary dimension, which is crucial to segment color images. 1
TRANSPORTATION DISTANCES ON THE CIRCLE
, 906
"... ABSTRACT. In this contribution, we study MongeKantorovich distances between discrete set of points on the unit circle S 1, when the ground distance between two points x and y on the circle is defined as c(x, y) = min(x − y,1 − x − y). We first prove that computing a MongeKantorovich distance ..."
Abstract

Cited by 4 (0 self)
 Add to MetaCart
(Show Context)
ABSTRACT. In this contribution, we study MongeKantorovich distances between discrete set of points on the unit circle S 1, when the ground distance between two points x and y on the circle is defined as c(x, y) = min(x − y,1 − x − y). We first prove that computing a MongeKantorovich distance between two given sets of pairwise different points boils down to cut the circle at a well chosen point and to compute the same distance on the real line. This result is then used to prove a formula on the Earth Mover’s Distance [3], which is a particular MongeKantorovich distance. This formula asserts that the Earth Mover’s Distance between two discrete circular normalized histograms f = (f[i])i=0,...,N−1 and g = (g[i])i=0,...,N−1 on N bins can be computed by (1) CEMD(f, g) = min k∈{0,...,N−1} ‖Fk − Gk‖1, where Fk and Gk are the cumulative histograms of f and g starting at the k th quantization bin. This formula is used in recent papers [1, 2] on the matching of local features between images, where the Earth Mover’s Distance is used to compare circular histograms of gradient orientations. 1.
SARSIFT: A SIFTLIKE ALGORITHM FOR SAR IMAGES
, 2013
"... The Scale Invariant Feature Transform (SIFT) algorithm is widely used in computer vision to match features between images or to localize and recognize objets. However, mostly because of speckle noise, it does not perform well on synthetic aperture radar (SAR) images. We present here an improvement ..."
Abstract

Cited by 2 (0 self)
 Add to MetaCart
The Scale Invariant Feature Transform (SIFT) algorithm is widely used in computer vision to match features between images or to localize and recognize objets. However, mostly because of speckle noise, it does not perform well on synthetic aperture radar (SAR) images. We present here an improvement of this algorithm for SAR images, named SARSIFT. A new gradient computation, yielding an orientation and a magnitude robust to speckle noise, is first introduced. It is then used to adapt several steps of the SIFT algorithm to SAR images. We study the improvement brought by this new algorithm, compared to existing approaches. We present an application of SARSIFT for the registration of SAR images in different configurations, especially with different incidence angles.
A contrario edge detection with edgelets
"... Abstract—Edge detection remains an active problem in the image processing community, because of the high complexity of natural images. In the last decade, Desolneux et al. proposed a novel detection approach, parameter free, based on the Helmhotz principle. Applied to the edge detection field, this ..."
Abstract

Cited by 2 (1 self)
 Add to MetaCart
(Show Context)
Abstract—Edge detection remains an active problem in the image processing community, because of the high complexity of natural images. In the last decade, Desolneux et al. proposed a novel detection approach, parameter free, based on the Helmhotz principle. Applied to the edge detection field, this means that observing a true edge in random and independent conditions is very unlikely, and then considered as meaningful. However, overdetection may occur, partly due to the use of a single pixelwise feature. In this paper, we propose to introduce higher level information in the a contrario framework, by computing several features along a set of connected pixels (an edgelet). Among the features, we introduce a shape prior, learned on a database. We propose to estimate online the a contrario distributions of the two other features, namely the gradient and the texture, by a MonteCarlo simulation approach. Experiments show that our method improves the original one, by decreasing the number of non relevant edges while preserving the true ones. I.
Spaceefficient approximation scheme for circular earth mover distance
 In LATIN
, 2012
"... Abstract. The Earth Mover Distance (EMD) between point sets A and B is the minimum cost of a bipartite matching between A and B. EMD is an important measure for estimating similarities between objects with quantifiable features and has important applications in several areas including computer visi ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
(Show Context)
Abstract. The Earth Mover Distance (EMD) between point sets A and B is the minimum cost of a bipartite matching between A and B. EMD is an important measure for estimating similarities between objects with quantifiable features and has important applications in several areas including computer vision. The streaming complexity of approximating EMD between point sets in a twodimensional discretized grid is an important open problem proposed in [8, 9]. We study the problem of approximating EMD in the streaming model, when points lie on a discretized circle. Computing the EMD in this setting has applications to computer vision [13] and can be seen as a special case of computing EMD in on a discretized grid. We achieve a (1 ± ε) approximation for EMD in Õ(ε−3) space, for every 0 < ε < 1. To our knowledge, this is the first streaming algorithm for a natural and widely applied EMD model that matches the space bound asked in [9]. 1
HAL manuscript No. 00399832 Transportation Distances on the Circle and Applications
, 2010
"... HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L’archive ouverte p ..."
Abstract
 Add to MetaCart
HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et a ̀ la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
AN ANALYSIS OF SCALESPACE SAMPLING IN SIFT
"... The most popular image matching algorithm SIFT, introduced by D. Lowe a decade ago, has proven to be sufficiently scale invariant to be used in numerous applications. In practice, however, scale invariance may be weakened by various sources of error. The density of the sampling of the Gaussian scal ..."
Abstract
 Add to MetaCart
The most popular image matching algorithm SIFT, introduced by D. Lowe a decade ago, has proven to be sufficiently scale invariant to be used in numerous applications. In practice, however, scale invariance may be weakened by various sources of error. The density of the sampling of the Gaussian scalespace and the level of blur in the input image are two of these sources. This article presents an empirical analysis of their impact on the extracted keypoints stability. We prove that SIFT is really scale and translation invariant only if the scalespace is significantly oversampled. We also demonstrate that the threshold on the difference of Gaussians value is inefficient for eliminating aliasing perturbations.
FASTMATCH: FAST AND ROBUST FEATURE MATCHING ON LARGE IMAGES
"... Today’s cameras produce images that often exceed 10 megapixels. Yet computing and matching local features for images of this size can easily take 20 seconds or more using optimized matching algorithms. This is much too slow for interactive applications and much too expensive for large scale image ..."
Abstract
 Add to MetaCart
Today’s cameras produce images that often exceed 10 megapixels. Yet computing and matching local features for images of this size can easily take 20 seconds or more using optimized matching algorithms. This is much too slow for interactive applications and much too expensive for large scale image operations. We introduce FastMatch, an algorithm designed to match large images efficiently without compromising matching accuracy. It derives its speed from only computing features in those parts of the image that can be confidently matched. FastMatch is an order of magnitude faster than the popular RatioMatch, yet often doubles matching precision for difficult image pairs. 1.