José L Alba-Castro | Universidade de Vigo (original) (raw)

Uploads

Papers by José L Alba-Castro

Proceedings of International Conference on Neural Networks (ICNN'97), 1997

ABSTRACT

Proceedings of International Conference on Neural Networks (ICNN'96)

ABSTRACT

2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems (BTAS)

In this work we thoroughly study the well-known face verification Resnet model in dlib’s library ... more In this work we thoroughly study the well-known face verification Resnet model in dlib’s library to uncover inner features related to soft biometrics attributes like gender, race and age. The study makes use of the t-SNE technique to understand the evolution of clustering through the pretrained network layers and reveals an interesting property of t-SNE to spot separability of clusters in the original space. The performance of simple classifiers for the secondary soft-biometrics tasks through the network reinforce the findings about t-SNE. This study is extensible to any model that maps the input classes into an embedded low-dimensional space that learned to cluster them in task-meaningful sets. We conclude that a state of the art face verification model can be easily leveraged to state of the art soft biometrics model without resorting to fine-tuning convolutional weights, which also allows reducing the model size and inference time.

1996 8th European Signal Processing Conference (EUSIPCO 1996), 1996

This paper presents a novel architecture based on a constructive algorithm that allows the networ... more This paper presents a novel architecture based on a constructive algorithm that allows the network to grow attending to both supervised and unsupervised criteria. The main goal is to end up with a set of discriminant functions able to solve a multi-class classification problem. The main difference with well-known NN-classificators lean on the fact that training is performed over labeled sets of patterns that we call high-level-structures (HLS). Every set contain patterns linked each other by some physical evidence, like neighbor pixels in a subimage or a time-sequence of frequency vectors in a speech utterance, but the membership of every individual pattern in the high-level-structure can not be so clear. This architecture has been tested on a number of artificial data sets and real data sets with very good results. We are now applying the algorithm to classification of real images drawn from the DataBase created for the ALINSPEC project.

Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205), 2001

This paper presents a face recognition system based on the use of image ridges and valleys as fac... more This paper presents a face recognition system based on the use of image ridges and valleys as face shape descriptors and a supervised Hausdorff based measure as similarity criteria. The proposed measure is designed to decrease the distance measured between images of the same subject. This measure is approximated using eigendecomposition in order to obtain a computationally efficient recognition method

Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205), 2001

Pattern Recognition in Information Systems, 2001

International Journal of Pattern Recognition and Artificial Intelligence, 2002

Hausdorff distance is a deformation tolerant measure between two sets of points. The main advanta... more Hausdorff distance is a deformation tolerant measure between two sets of points. The main advantage of this measure is that it does not need an explicit correspondence between the points of the two sets. This paper presents the application to automatic face recognition of a novel supervised Hausdorff-based measure. This measure is designed to minimize the distance between sets of the same class (subject) and at the same time maximize the distance of sets between different classes.

$Research paper thumbnail of Reducing the complexity in a fractal-based image coder$

Proceedings of IEEE 6th Digital Signal Processing Workshop

... FJ Gonzdez, D. Docampo and JL Alba Dept. ... A dimensionality-reduced subvec-tor is obtained ... more ... FJ Gonzdez, D. Docampo and JL Alba Dept. ... A dimensionality-reduced subvec-tor is obtained by extracting the coefficients of impor-tance, from a human visual point of view. Then, a candidate list of the nearest neighbors, in the trans-form domain, is stored. ...

Signal Processing, 1999

In this paper a method to automatically generate a Gaussian mixture classifier is presented. The ... more In this paper a method to automatically generate a Gaussian mixture classifier is presented. The growing process is based on the iterative addition of Gaussian nodes. Each iteration takes place in two sequential steps: first, using the EM algorithm, we maximize the likelihood of the data under the current configuration of the classifier; then, a new Gaussian node is added

Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429), 2003

Computer Vision and Image Understanding, Nov 1, 2020

Proceedings of SPIE, Nov 14, 2001

In this article we present an hybrid SOM+PCA approach for face identification that is based on se... more In this article we present an hybrid SOM+PCA approach for face identification that is based on separating shape and texture information. Shape will be processed by a modified Hausdorff distance SOM and texture processing relies on a modular PCA. In most successfully view-based recognition systems, shape and texture are jointly used to statistically model a linear or piecewise linear subspace that optimally explains the face space for a specific database. Our work is aimed to separate the influence that variance in face shape stamps on the set of eigenfaces in the classical PCA decomposition. In this sense we search for a more efficiently coded face-vector for identification. The ultimate goal consist of finding a non-linear transformation invariant to gesture changes and, in a larger extent, to pose changes. The first part of this paper is dedicated to the shape processor of the system, that is based on a novel shape-based Self Organizing Map, and the second part deals with the subspace PCA decomposition that relies on the SOM clustering. Results are reported by comparing face identification between PCA and the SOM-PCA approach.

Esta tesis propone nuevas soluciones constructivas al problema del entrenamiento de una red neuro... more Esta tesis propone nuevas soluciones constructivas al problema del entrenamiento de una red neuronal hibrida aplicada a la clasificacion de patrones. El trabajo se plantea desde dos perspectivas distintas: una conexionista utilizando tecnicas fundamentadas en algoritmos propios del campo de las redes neuronales, y otra probabilistica, con soluciones mas afines al campo de la estadistica. Cada uno de estos dos planteamientos deriva en una arquitectura constructiva que es capaz de generar un clasificador de prestaciones comparables a modelos ampliamente difundidos, como el perceptron multicapa (mlp), las redes con funciones de base radial (rbf) o los clasificadores basados en la regla de los k vecinos mas proximos (k-nn). Ambas arquitecturas constructivas incorporan instrumentos para controlar la complejidad del modelo con el proposito de mantener una generalizacion adecuada. El control de la complejidad sirve asimismo, como una forma de detener el proceso de entrenamiento. A lo largo de la tesis se muestra un conjunto de experimentos con datos generados artificialmente que reflejan las prestaciones de ambas redes neuronales y permiten evaluar la influencia, en el crecimiento y en la tasa de error, de los distintos parametros incorporados en los algoritmos. Las pruebas realizadas con datos procedentes de aplicaciones reales corroboran las conclusiones extraidas a partir de las pruebas artificiales y de los teoricos que fundamentan los algoritmos.