Jaume Amores - Academia.edu (original) (raw)

Papers by Jaume Amores

Multiple Instance Learning (MIL) has become a hot topic and many different algorithms have been p... more Multiple Instance Learning (MIL) has become a hot topic and many different algorithms have been proposed in the last years. Despite this fact, there is a lack of comparative studies that shed light into the characteristics of the different methods and their behavior in different scenarios. In this paper we provide such an analysis. We include methods from different families,

IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007

We present a novel approach for retrieval of object categories based on a novel type of image rep... more We present a novel approach for retrieval of object categories based on a novel type of image representation: the Generalized Correlogram (GC). In our image representation, the object is described as a constellation of GCs where each one encodes information about some local part and the spatial relations from this part to others (i.e. the part's context). We show how such a representation can be used with fast procedures that learn the object category with weak supervision and efficiently match the model of the object against large collections of images. In the learning stage, we show that by integrating our representation with Boosting the system is able to obtain a compact model that is represented by very few features, where each feature conveys key properties about the object's parts and their spatial arrangement. In the matching step, we propose direct procedures that exploit our representation for efficiently considering spatial coherence between the matching of local parts. Combined with an appropriate data organization such as Inverted Files, we show that thousands of images can be evaluated efficiently. The framework has been applied to the standard CALTECH database with seven object categories and clutter, and we show that our results are favorably compared against state-of-the-art methods in both computational cost and accuracy.

International Journal of Intelligent Systems, 2005

We present a content-based image retrieval system of medical images of bodies with high elas-tici... more We present a content-based image retrieval system of medical images of bodies with high elas-ticity, that is, high inter-and intrasubject variability in shape. The system is based on a rich feature space that is able to describe all the relevant aspects of an image, including local, global, and contextual information. For including relative spatial relations between the structures (i. e., contextual information) we do not need to obtain very accurate segmentations of the image, in contrast to the majority of methods employed for this kind of description. We also obtain invari-ance to spatial deformations of the same type of object along different instances. This is achieved by applying efficient registrations of the images before their comparison. The incorporation of all these components represents an innovative and powerful way of comparing and retrieving medical images. Validated results are reported on a database of 168 intravascular ultrasound images, showing the appropriateness of our approach for images of such high complexity. ? 2005 Wiley Periodicals, Inc.

In this paper, we present a general guideline to establish the relation between a distribution mo... more In this paper, we present a general guideline to establish the relation between a distribution model and its corresponding similarity estimation. A rich set of distance metrics, such as harmonic distance and geometric distance, is derived according to Maximum Likelihood theory. These metrics can provide a more accurate feature model than the conventional Euclidean distance (SSD) and Manhattan distance (SAD). Because the feature elements are from heterogeneous sources and may have different influence on similarity estimation, the assumption of single isotropic distribution model is often inappropriate. We propose a novel boosted distance metric that not only finds the best distance metric that fits the distribution of the underlying elements but also selects the most important feature elements with respect to similarity. We experiment with different distance metrics for similarity estimation and compute the accuracy of different methods in two applications: stereo matching and motion tracking in video sequences. The boosted distance metric is tested on fifteen benchmark data sets from the UCI repository and two image retrieval applications. In all the experiments, robust results are obtained based on the proposed methods.

Object-class recognition is one of the most challenging fields of pattern recognition and compute... more Object-class recognition is one of the most challenging fields of pattern recognition and computer vision. Currently, most authors represent an object as a collection of parts and their mutual spatial relations. Therefore, two types of information are extracted: local information describing each part, and contextual information describing the (spatial) context of the part, i.e. the spatial relations between the rest of the parts and the current one. We define a generalized correlogram descriptor and represent the object as a constellation of such generalized correlograms. Using this representation, both local and contextual information are gathered into the same feature space. We take advantage of this representation in the learning stage, by using a feature selection with boosting that learns both types of information simultaneously and very efficiently. Simultaneously learning both types of information proves to be a faster approach than dealing with them separately. Our method is compared with state-of-the-art object-class recognition systems by evaluating both the accuracy and the cost of the methods.

... We can use either the 2-D correlograms or the 1-D correlograms. We have seen small difference... more ... We can use either the 2-D correlograms or the 1-D correlograms. We have seen small difference between them in the results because the 2-D correl-ograms perform as well as the 1-D correlograms when the shapes are already aligned. ...

We present a registration and retrieval algorithm of medical images. Our algorithm is oriented in... more We present a registration and retrieval algorithm of medical images. Our algorithm is oriented in a general fashion towards gray level medical images of non-rigid bodies such as coronary vessels, where object shape information provide poor information. We use rich descriptors based on both local and global (contextual) information, and at the same time we use a cooperative-iterative strategy in order to get a good set of correspondences as well as a good final transformation. We focus on a novel application of registration of medical images: registration of IVUS, a promising technique of analyzing the coronary vessels.

Pattern Recognition Letters, 2005

In medical imaging, comparing and retrieving objects is non-trivial because of the high variabili... more In medical imaging, comparing and retrieving objects is non-trivial because of the high variability in shape and appearance. Such variety leads to poor performance of retrieval algorithms only based on local or global descriptors (shape, color, texture). In this article, we propose a context-based framework for medical image retrieval on the grounds of a global object context based on the mutual positions of local descriptors. This characterization is incorporated into a fast non-rigid registration process to provide invariance against elastic transformations. We apply our method to a complex domain of images-retrieval of intravascular ultrasound images according to vessel morphology. Final results are very encouraging.

We present a novel approach for fast object class recognition incorporating contextual informatio... more We present a novel approach for fast object class recognition incorporating contextual information into boosting. The object is represented as a constellation of generalized correlograms that integrate both information of local parts and their spatial relations. Incorporating the spatial relations into our constellation of descriptors, we show that an exhaustive search for the best matching can be avoided. Combining the contextual descriptors with boosting, the system simultaneously learns the information that characterize each part of the object along with their characteristic mutual spatial relations. The proposed framework includes a matching step between homologous parts in the training set, and learning the spatial pattern after matching. In the matching part two approaches are provided: a supervised algorithm and an unsupervised one. Our results are favorably compared against state-of-the-art results.

International Journal of Intelligent Systems

Abstract: We present a content-based image retrieval system of medical images ofbodies with high ... more Abstract: We present a content-based image retrieval system of medical images ofbodies with high elasticity, i.e. high inter and intra subject variability in shape. Thesystem is based on a rich feature space which is able to describe all the relevant aspectsof an image, including local, global and contextual information. For includingrelative spatial relations between the structures (i.e. contextual information) we do notneed to obtain very accurate segmentations of the image, in contrast to...

Distance metric is widely used in similarity estimation. In this paper we find that the most popu... more Distance metric is widely used in similarity estimation. In this paper we find that the most popular Euclidean and Manhattan distance may not be suitable for all data distributions. A general guideline to establish the relation between a distribution model and its corresponding similarity estimation is proposed. Based on Maximum Likelihood theory, we propose new distance metrics, such as harmonic distance and geometric distance. Because the feature elements may be from heterogeneous sources and usually have different influence on similarity estimation, it is inappropriate to model the distribution as isotropic. We propose a novel boosted distance metric that not only finds the best distance metric that fits the distribution of the underlying elements but also selects the most important feature elements with respect to similarity. The boosted distance metric is tested on fifteen benchmark data sets from the UCI repository and two image retrieval applications. In all the experiments, robust results are obtained based on the proposed methods.

Information and Computation/information and Control, 2006

In this work we introduce a new distance estimation technique by boosting and we apply it to the ... more In this work we introduce a new distance estimation technique by boosting and we apply it to the K-Nearest Neighbor Classifier (K-NN). Instead of applying AdaBoost to a typical classification problem, we use it for learning a distance function and the resulting distance is used into K-NN. The proposed method (Boosted Distance with Nearest Neighbor) outperforms the AdaBoost classifier when the training set is small. It also outperforms the K-NN classifier used with several different distances and the distances obtained with other estimation methods such as Relevant Component Analysis (RCA) [Duda, R.O., Hart, P.E., Stock, D.G., 2001. Pattern Classification, John Wiley and Sons Inc.]. Furthermore, our distance estimation performs dimension-reduction, being much more efficient in terms of classification accuracy than classical techniques such as PCA, LDA, and NDA. The method has been thoroughly tested on 13 standard databases from the UCI repository, a standard gender recognition database and the MNIST database.

We present a new framework for characterizing and retrieving objects in cluttered scenes. This CB... more We present a new framework for characterizing and retrieving objects in cluttered scenes. This CBIR system is based on a new representation describing every object taking into account the local properties of its parts and their mutual spatial relations, without relying on accurate segmentation. For this purpose, a new multi-dimensional histogram is used that measures the joint distribution of local properties and relative spatial positions. Instead of using a single descriptor for all the image, we represent the image by a set of histograms covering the object from different perspectives. We integrate this representation in a whole framework which has two stages. The first one is to allow an efficient retrieval based on the geometric properties (shape) of objects in images with clutter. This is achieved by i) using a contextual descriptor that incorporates the distribution of local structures, and ii) taking a proper distance that disregards the clutter of the images. At a second stage, we introduce a more discriminative descriptor that characterizes the parts of the objects by their color and their local structure. By using relevant-feedback and boosting as a feature selection algorithm, the system is able to learn simultaneously the information that characterize each part of the object along with their mutual spatial relations. Results are reported on two known databases and are quantitatively compared to other successful approaches.

Advances in Imaging and Electron Physics, 2006

In this work, we present a general guideline to establish the relation between a distribution mod... more In this work, we present a general guideline to establish the relation between a distribution model and its corresponding similarity estimation. A rich set of distance metrics, such as Harmonic distance and Geometric distance, is derived according to Maximum Likelihood theory.

We present a content-based image retrieval system of medical images of bodies with high elas- tic... more We present a content-based image retrieval system of medical images of bodies with high elas- ticity, that is, high inter- and intrasubject variability in shape. The system is based on a rich feature space that is able to describe all the relevant aspects of an image, including local, global, and contextual information. For including relative spatial relations between the structures

Pattern Recognition Letters, 2003

IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008

In this paper, we present a general guideline to find a better distance measure for similarity es... more In this paper, we present a general guideline to find a better distance measure for similarity estimation based on statistical analysis of distribution models and distance functions. A new set of distance measures are derived from the harmonic distance, the geometric distance, and their generalized variants according to the Maximum Likelihood theory. These measures can provide a more accurate feature model than the classical euclidean and Manhattan distances. We also find that the feature elements are often from heterogeneous sources that may have different influence on similarity estimation. Therefore, the assumption of single isotropic distribution model is often inappropriate. To alleviate this problem, we use a boosted distance measure framework that finds multiple distance measures, which fit the distribution of selected feature elements best for accurate similarity estimation. The new distance measures for similarity estimation are tested on two applications: stereo matching and motion tracking in video sequences. The performance of boosted distance measure is further evaluated on several benchmark data sets from the UCI repository and two image retrieval applications. In all the experiments, robust results are obtained based on the proposed methods.

This paper presents an efficient object-class recognition approach based on a new type of image d... more This paper presents an efficient object-class recognition approach based on a new type of image descriptor: the Class-Specific Binary Correlogram (CSBC). In our representation, the image is described by a collection of CSBCs, where each one encodes the spatial distribution of class-specific features around a particular reference point. This representation is obtained by first performing an automatic selection of class-specific features from a vocabulary, and then extracting collections of binary correlograms that encode, at the same time, detected object parts and their spatial distribution around multiple points of the image. Our descriptors live in high-dimensional spaces (in the order of 10K dimensions), but they are very sparse. We show that efficient learning and matching procedures can be obtained for such a representation if we use, first, fast feature selection techniques specific for binary features, and then Boosting integrated with an appropriate Inverted File data organization. The proposed strategy works with weak supervision, outperforms state-of-the-art bag-of-feature methods, and it is more accurate and computationally more efficient than well-known geometrical-based methods, including our previous work on Generalized Correlograms (GCs) .