1,721,339 research outputs found
Weakly Supervised Learning of Metric Aggregations for Deformable Image Registration
No full textDeformable registration has been one of the pillars of biomedical image computing. Conventional approaches refer to the definition of a similarity criterion that, once endowed with a deformation model and a smoothness constraint, determines the optimal transformation to align two given images. The definition of this metric function is among the most critical aspects of the registration process. We argue that incorporating semantic information (in the form of anatomical segmentation maps) into the registration process will further improve the accuracy of the results. In this paper, we propose a novel weakly supervised approach to learn domain-specific aggregations of conventional metrics using anatomical segmentations. This combination is learned using latent structured support vector machines. The learned matching criterion is integrated within a metric-free optimization framework based on graphical models, resulting in a multi-metric algorithm endowed with a spatially varying similarity metric function conditioned on the anatomical structures. We provide extensive evaluation on three different datasets of CT and MRI images, showing that learned multi-metric registration outperforms single-metric approaches based on conventional similarity measures.Fil: Ferrante, Enzo. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Santa Fe. Instituto de Investigación en Señales, Sistemas e Inteligencia Computacional. Universidad Nacional del Litoral. Facultad de Ingeniería y Ciencias Hídricas. Instituto de Investigación en Señales, Sistemas e Inteligencia Computacional; ArgentinaFil: Dokania, Puneet Kumar. University of Oxford; Reino UnidoFil: Silva, Rafael Marini. Centre de Vision Numérique; . Therapanacea;Fil: Paragios, Nikos. Therapanacea; . Centre de Vision Numérique
Registration of Structures in Arbitrary Dimensions: Implicit Representations, Mutual Information & Free form Deformations
No full textRegistration is a core component in various applications of imaging and vision. While simple cases refer to the registration of clouds of points, a strong need exists for shape, image and volume alignment. In this paper, we propose a novel global-to-local registration method that integrates statistical and variational techniques. Registration is considered in an implicit higher dimensional space. The powerful space of distance transforms of arbitrary metric is used as an embedding function. Mutual information can support various motion models and is considered to perform global registration. A B-Spline approximation of grid is used within a free-from deformation criterion to recover a (complementary to the global) dense registration field that is continuous and guarantees one-to-one mapping. Such framework exhibits robustness and can cope in an efficient manner with important local deformations. 2D/3D shapes are used to demonstrate the potentials of the proposed technique.Technical report DCS-TR-52
Evaluations on multi-scale camera networks for precise and geo-accurate reconstructions from aerial and terrestrial images with user guidance
No full textDuring the last decades photogrammetric computer vision systems have been well established in scien- tific and commercial applications. Recent developments in image-based 3D reconstruction systems have resulted in an easy way of creating realistic, visually appealing and accurate 3D models. We present a fully automated processing pipeline for metric and geo-accurate 3D reconstructions of complex geome- tries supported by an online feedback method for user guidance during image acquisition. Our approach is suited for seamlessly matching and integrating images with different scales, from different view points (aerial and terrestrial), and with different cameras into one single reconstruction. We evaluate our ap- proach based on different datasets for applications in mining, archaeology and urban environments and thus demonstrate the flexibility and high accuracy of our approach. Our evaluation includes accuracy related analyses investigating camera self-calibration, georegistration and camera network configuration
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Coupled Gaussian Process Regression for Pose-Invariant Facial Expression Recognition
Full text linkWe present a novel framework for the recognition of facial expressions at arbitrary poses that is based on 2D geometric features. We address the problem by first mapping the 2D locations of landmark points of facial expressions in non-frontal poses to the corresponding locations in the frontal pose. Then, recognition of the expressions is performed by using any state-of-the-art facial expression recognition method (in our case, multi-class SVM). To learn the mappings that achieve pose normalization, we use a novel Gaussian Process Regression (GPR) model which we name Coupled Gaussian Process Regression (CGPR) model. Instead of learning single GPR model for all target pairs of poses at once, or learning one GPR model per target pair of poses independently of other pairs of poses, we propose CGPR model, which also models the couplings between the GPR models learned independently per target pairs of poses. To the best of our knowledge, the proposed method is the first one satisfying all: (i) being face-shape-model-free, (ii) handling expressive faces in the range from −45◦ to +45◦ pan rotation and from −30◦ to +30◦ tilt rotation, and (iii) performing accurately for continuous head pose despite the fact that the training was conducted only on a set of discrete poses
Localizing Objects While Learning Their Appearance
Full text linkLearning a new object class from cluttered training images is very challenging when the location of object instances is unknown. Previous works generally require objects covering a large portion of the images. We present a novel approach that can cope with extensive clutter as well as large scale and appearance variations between object instances. To make this possible we propose a conditional random field that starts from generic knowledge and then progressively adapts to the new class. Our approach simultaneously localizes object instances while learning an appearance model specific for the class. We demonstrate this on the challenging Pascal VOC 2007 dataset. Furthermore, our method enables to train any state-of-the-art object detector in a weakly supervised fashion, although it would normally require object location annotations
Manifold Valued Statistics, Exact Principal Geodesic Analysis and the Effect of Linear Approximations
No full textManifolds are widely used to model non-linearity arising in a range of computer vision applications. This paper treats statistics on manifolds and the loss of accuracy occurring when linearizing the manifold prior to performing statistical operations. Using recent advances in manifold computations, we present a comparison between the non-linear analog of Principal Component Analysis, Principal Geodesic Analysis, in its linearized form and its exact counterpart that uses true intrinsic distances. We give examples of datasets for which the linearized version provides good approximations and for which it does not. Indicators for the dierences between the two versions are then developed and applied to two examples of manifold valued data: outlines of vertebrae from a study of vertebral fractures and spacial coordinates of human skeleton end-eectors acquired using a stereo camera and tracking software.Manifolds are widely used to model non-linearity arising in a range of computer vision applications. This paper treats statistics on manifolds and the loss of accuracy occurring when linearizing the manifold prior to performing statistical operations. Using recent advances in manifold computations, we present a comparison between the non-linear analog of Principal Component Analysis, Principal Geodesic Analysis, in its linearized form and its exact counterpart that uses true intrinsic distances. We give examples of datasets for which the linearized version provides good approximations and for which it does not. Indicators for the differences between the two versions are then developed and applied to two examples of manifold valued data: outlines of vertebrae from a study of vertebral fractures and spacial coordinates of human skeleton end-effectors acquired using a stereo camera and tracking software.<br/
Learning Pre-attentive Driving Behaviour from Holistic Visual Features
No full textThe aim of this paper is to learn driving behaviour by associating the actions recorded from a human driver with pre-attentive visual input, implemented using holistic image features (GIST). All images are labelled according to a number of driving–relevant contextual classes (eg, road type, junction) and the driver’s actions (eg, braking, accelerating, steering) are recorded. The association between visual context and the driving data is learnt by Boosting decision stumps, that serve as input dimension selectors. Moreover, we propose a novel formulation of GIST features that lead to an improved performance for action prediction. The areas of the visual scenes that contribute to activation or inhibition of the predictors is shown by drawing activation maps for all learnt actions. We show good performance not only for detecting driving–relevant contextual labels, but also for predicting the driver’s actions. The classifier’s false positives and the associated activation maps can be used to focus attention and further learning on the uncommon and difficult situations
- …
