Content uploaded by Manuel Günther

Author content

All content in this area was uploaded by Manuel Günther on Oct 06, 2017

Content may be subject to copyright.

A preview of the PDF is not available

Content uploaded by Manuel Günther

Author content

All content in this area was uploaded by Manuel Günther on Oct 06, 2017

Content may be subject to copyright.

A preview of the PDF is not available

... where J = ν max ζ max is the cardinality of the Gabor filter subset generated with the Γ parameterization. Besides this common set of parameter values, the present work employs two additional ones initially proposed by Günther [36] to enable gray level and color image reconstruction from model graphs. One of them is the extended parameterization Γ {g} , which includes Gabor filters that capture supplementary high and low frequency information, The other one provides color information using the YUV color space. ...

... The absolute values or magnitudes from these feature descriptors remain similar for small displacements in their offset location [36] modeling the behavior of complex cells [99] from the primary visual cortex. 2 The convolution of a Gabor filter with an image is used to estimate the magnitude of existing frequencies that have a similar wavelength and orientation. ...

... Westphal [94] reports smooth changes of values with fairly wide maxima when applying this similarity function to a single Gabor jet taken from the center of a probe image with respect to translation, scale, and rotation. Günther [36] extends this analysis to multiple scales and rotation angles to measure the impact of image resolution variations on Gabor jet similarities. Its empirical results indicate that Gabor jet similarities are very sensitive to scale and rotation differences in their underlying image area. ...

Using distributed representations of objects enables artificial systems to be more versatile regarding inter- and intra-category variability, improving the appearance-based modeling of visual object understanding. They are built on the hypothesis that object models are structured dynamically using relatively invariant patches of information arranged in visual dictionaries, which can be shared across objects from the same category. However, implementing distributed representations efficiently to support the complexity of invariant object recognition and categorization remains a research problem of outstanding significance for the biological, the psychological, and the computational approach to understanding visual perception. The present work focuses on solutions driven by top-down object knowledge. It is motivated by the idea that, equipped with sensors and processing mechanisms from the neural pathways serving visual perception, biological systems are able to define efficient measures of similarities between properties observed in objects and use these relationships to form natural clusters of object parts that share equivalent ones. Based on the comparison of stimulus-response signatures from these object-to-memory mappings, biological systems are able to identify objects and their kinds. The present work combines biologically inspired mathematical models to develop memory frameworks for artificial systems, where these invariant patches are represented with regular-shaped graphs, whose nodes are labeled with elementary features that capture texture information from object images. It also applies unsupervised clustering techniques to these graph image features to corroborate the existence of natural clusters within their data distribution and determine their composition. The properties of such computational theory include self-organization and intelligent matching of these graph image features based on the similarity and co-occurrence of their captured texture information. The performance to model invariant object recognition and categorization of feature-based artificial systems equipped with each of the developed memory frameworks is validated applying standard methodologies to well-known image libraries found in literature. Additionally, these artificial systems are cross-compared with state-of-the-art alternative solutions. In conclusion, the findings of the present work convey implications for strategies and experimental paradigms to analyze human object memory as well as technical applications for robotics and computer vision.

... Gabor wavelets are widely used for face recognition [10,2,11,1] and as models for processing in the primary visual cortex. The complex-valued Gabor responses are split up into an amplitude and a phase, which is called Gabor phase. ...

... Gabor wavelet responses at single locations of facial images are collected into Gabor jets [10], which are extracted at several offset positions and assembled into a Gabor graph G. Often, face graphs with nodes at facial landmark positions are utilized [10,2], but in the present paper we use rectangular grid graphs, which have shown to be at least as expedient [1,8]. ...

... The texture contents of Gabor graphs can be displayed by reconstruction [2]. Exemplary reconstructions of grid graphs are shown in Figure 1 To show the significance of Gabor phases for face recognition, absolute values and phase values from graphs G A and G B are merged into G A,B and G B,A , where each Gabor jet holds amplitudes of G A and phases of G B or vice versa. ...

We analyze the relative relevance of Gabor amplitudes and phases for face recognition. We propose an algorithm to reliably estimate offset point disparities from phase differences and show that disparity-corrected Gabor phase differences are well suited for face recognition in difficult lighting conditions. The method reaches 74.8% recognition rate on the Lighting set of the CAS-PEAL database and 35.7% verification rate on experiment 2.4 of the FRGC database.

... 16,25 Recently, the data format has been applied successfully to the classification of different syndromes, which influence the facial appearance, from static facial images. 5,10,12,15,18,21,23 To better interpret which kind of features are contained in the Gabor graphs and for inspection by the clinician, it is important to visualize Gabor graphs by reconstructing images from them. Daugman 6,7 used a neural network to approximate the expansion coefficients of a Gabor wavelet transform in full resolution. ...

... The calculation of the denominatorČ ψ 2, 22 in Equation (3.5) can be deduced from wavelet theory. 12 TheČ min parameter, which confinesČ not to vanish, has to be lower than the sum of squared Gabor wavelets in frequency domain, when ω is inside of the covered sub-band, but should not be too small. We chooseČ min = 0.25. ...

Graphs labeled with complex-valued Gabor jets are one of the important data formats for face recognition and the classification of facial images into medically relevant classes like genetic syndromes. We here present an interpolation rule and an iterative algorithm for the reconstruction of images from these graphs. This is especially important if graphs have been manipulated for information processing. One such manipulation is averaging the graphs of a single syndrome, another one building a composite face from the features of various individuals. In reconstructions of averaged graphs of genetic syndromes, the patients' identities are suppressed, while the properties of the syndromes are emphasized. These reconstructions from average graphs have a much better quality than averaged images.

... Figura 3: Janelas de tempo-frequência: (a) trem de impulsos, (b) Transformada de Fourier e (c) Transformada Wavelet. Fonte: [11] Conforme exposto por Günther [13], pode-se definir uma wavelet mãe por: ...

Resumo-Alvo de diversas pesquisas, a estimação de posicio-namento de cabeça tem por objetivo determinar a posição de um indivíduo em uma determinada cena. Estratégias de análise de vídeo para estimação de posição de cabeça que funcionem com webcams convencionais têm um papel central em sistemas de rastreamento ocular de baixo-custo. Neste trabalho, os autores comparam três classificadores (baseados em kernel principal component analysis, KPCA; linear discriminant analysis, LDA; e support vector classification, SVC) para estimar a posição de cabeça em uma base de imagens contendo poses discretas. A extração de características de entrada dos classificadores utiliza transformada wavelet de Gabor, cujos aspectos conceituais básicos são apresentados no texto. Testes utilizando uma base de dados disponível publicamente na internet foram realizados. Os resultados mostram que os classificadores possuem acurácia superior a 70% para número total de classes não superior a 15. À medida em que o número de classes aumenta, o desempenho dos classificadores deteriora significativamente: para 93 classes, por exemplo, as acurácias medidas foram de 34%, 33% e 43% para os classificadores baseados em KPCA, LDA, e SVC, respectivamente.

... Following a general trend (e.g., [7]), image and model pixels are labelled as jets ) , ( y x of Gabor wavelets, using as convolution kernel ...

... Following a general trend (e.g., [7]), image and model pixels are labelled as jets ) , ( y x of Gabor wavelets, using as convolution kernel ...

A simple model of MNIST handwritten digit recognition is presented here. The model is an adaptation of a previous theory of face recognition. It realizes translation and rotation invariance in a principled way instead of being based on extensive learning from large masses of sample data. The presented recognition rates fall short of other publications, but due to its inspectability and conceptual and numerical simplicity, our system commends itself as a basis for further development.

... Following a general trend (e.g., [7]), image and model pixels are labelled as jets ) , ( y x of Gabor wavelets, using as convolution kernel ...

A simple model of MNIST handwritten digit recognition is presented here. The model is an adaptation of a previous theory of face recognition. It realizes translation and rotation invariance in a principled way instead of being based on extensive learning from large masses of sample data. The presented recognition rates fall short of other publications, but due to its inspectability and conceptual and numerical simplicity, our system commends itself as a basis for further development.

... The idea of the Graphs algorithm relies on a Gabor wavelet transform [56,57]. The preprocessed image is transformed using a family of j = 1, . . . ...

One important type of biometric authentication is face recognition
, a research area of high popularity with a wide spectrum of approaches that have been proposed in the last few decades. The majority of existing approaches are conceived for or evaluated on constrained still images. However, more recently research interests have shifted toward unconstrained “in-the-wild
” still images and videos. To some extent, current state-of-the-art systems are able to cope with variability due to pose, illumination, expression, and size, which represent the challenges in unconstrained face recognition. To date, only few attempts have addressed the problem of face recognition in mobile environment
, where high degradation is present during both data acquisition and transmission. This book chapter deals with face recognition in mobile and other challenging environments, where both still images and video sequences are examined. We provide an experimental study of one commercial off-the-shelf (COTS) and four recent open-source face recognition algorithms
, including color-based linear discriminant analysis (LDA)
, local Gabor binary pattern histogram sequences (LGBPHSs)
, Gabor grid graphs
, and intersession variability (ISV) modeling
. Experiments are performed
on several freely available challenging still image and video face databases, including one mobile database, always following the evaluation protocols that are attached to the databases. Finally, we supply an easily extensible open-source toolbox to rerun all the experiments, which includes the modeling techniques, the evaluation protocols, and the metrics used in the experiments and provides a detailed description on how to regenerate the results.

... The analysis of texture relies on the calculation of similarity functions for Gabor jets at the nodes. A Gabor jet is a vector that describes the image's texture information around the node [ 6 ] . The analysis of geometry is based on the comparison of the distances between nodes. ...

Objective:
Cushing’s syndrome causes considerable harm to the body if left untreated, yet often remains undiagnosed for prolonged periods of time. In this study we aimed to test whether face classification software might help in discriminating patients with Cushing’s syndrome from healthy controls.
Design:
Diagnostic study.
Patients:
Using a regular digital camera, we took frontal and profile pictures of 20 female patients with Cushing’s syndrome and 40 sex- and age-matched controls.
Measurements:
Semi-automatic analysis of the pictures was performed by comparing texture and geometry within a grid of nodes placed on the pictures. The leave-one-out cross-validation method was employed to classify subjects by the software.
Results:
The software correctly classified 85.0% of patients and 95.0% of controls, resulting in a total classification accuracy of 91.7%.
Conclusions:
In this preliminary analysis we found a good classification accuracy of Cushing’s syndrome by face classification software. Testing accuracy is comparable to that of currently employed screening tests.

We present a system for recognizing human faces from single images out of a large database containing one image per person. Faces are represented by labeled graphs, based on a Gabor wavelet transform. Image graphs of new faces are extracted by an elastic graph matching process and can be compared by a simple similarity function. The system differs from the preceding one in three respects. Phase information is used for accurate node positioning. Object-adapted graphs are used to handle large rotations in depth. Image graph extraction is based on a novel data structure, the bunch graph, which is constructed from a small set of sample image graphs.

This paper describes a visual object detection framework that is capable of processing images extremely rapidly while achieving high detection rates. There are three key contributions. The first is the introduction of a new image representation called the "Integral Image" which allows the features used by our detector to be computed very quickly. The second is a learning algorithm, based on AdaBoost, which selects a small number of critical visual features and yields extremely efficient number of critical visual features and yields extremely efﬁcient classiﬁers [6]. The third contribution is a method for combining classiﬁers in a “cascade” which allows background regions of the image to be quickly discarded while spending more computation on promising object-like regions. A set of experiments in the domain of face detection are presented. The system yields face detection performace comparable to the best previous systems [18, 13, 16, 12, 1]. Implemented on a conventional desktop, face detection proceeds at 15 frames per second.

This work covers three mathematical analysis domains: Continuous Wavelet Transform, Fourier Transform and Clifford Analysis, and consists of three parts in which these domains interact. The one-dimensional continuous wavelet transform (CWT) is a successful tool for signal and image analysis, with applications in mathematics, physics and engineering. Higher dimensional CWTs typically originate as tensor products of one-dimensional phenomena. Clifford analysis offers a natural generalization to higher dimension of the theory of holomorphic functions in the complex plane. The generalized holomorphic functions, known as monogenic functions, are null-solutions of the so-called Dirac operator, a first order rotationally invariant differential operator factorizing the Laplacian in higher dimensions. This factorization of the Laplace operator establishes a special relationship between monogenic functions and harmonic functions of several variables, in that the properties of monogenic functions constitute a refinement of those of harmonic functions. An intrinsic feature of Clifford analysis is that it encompasses all dimensions at once, as opposed to the usual tensorial approaches. This true multi-dimensional nature allows for a very specific construction of higher dimensional wavelets and the development of the corresponding CWT-theory, based on generalizations to higher dimension of classical orthogonal polynomials on the real line. In Part I this wavelet construction procedure is presented within the usual, orthogonal Clifford analysis framework, while in Part III we generalize it to the metric dependent setting of Clifford analysis. The latter gives rise to so-called anisotropic Clifford-wavelets which are adaptable to preferential, not necessarily orthogonal, directions in the signals or textures to be analysed. The central topic of Part II is the development of a new multi-dimensional Fourier transform in the framework of Clifford analysis, the so-called Clifford-Fourier transform. It arises as a theoretical construct quite naturally in the spirit of the above mentioned refinement of harmonic functions by monogenic ones.