In this work, a method is presented for the continuous estimation of pain intensity based on fusion of bio-physiological and video features. Furthermore, a method is proposed for the adaptation of the system to unknown test persons based... more
In this work, a method is presented for the continuous estimation of pain intensity based on fusion of bio-physiological and video features. Furthermore, a method is proposed for the adaptation of the system to unknown test persons based on unlabeled data. First, an analysis is presented that shows which modalities and feature sets are suited best for the task of recognizing pain levels in a person-independent setting. For this, a large set of features is extracted from the available bio-physiological channels (ECG, EMG and skin conductivity) and the video stream. We then propose a method to learn the confidence of a regression system using a multi-stage ensemble classifier. Based on the outcome of the classifier, which is realized by a neural network, confident samples are selected by the adaptation procedure. In various experiments, we show that the algorithm is able to detect highly confident samples which can be used to improve the overall performance. We furthermore discuss the current limitations of automatic pain intensity estimation—in light of the presented approach and beyond.
Research Interests:
In this work, a method is proposed to train Support Vector Machines in parallel. The dierence to other parallel implementations is that the problem is decomposed into hierarchically connected nodes and that each node does not have to... more
In this work, a method is proposed to train Support Vector Machines in parallel. The dierence to other parallel implementations is that the problem is decomposed into hierarchically connected nodes and that each node does not have to fully optimize its local problem. Instead Lagrange multipliers are ltered and transferred between nodes during runtime, with important ones ascending and unimportant ones descending inside the architecture. Experimental validation demonstrates the advan- tages in terms of speed in comparison to other approaches.
Research Interests:
... 7 Distributed Architecture for Speech-Controlled Systems Based on Associative Memories ZöhreKara Kayikci1), Dmitry Zaykovskiy1), Heiner Markert1 ... The mel filter block divides this spectrum into 23 equidistant, weighted,... more
... 7 Distributed Architecture for Speech-Controlled Systems Based on Associative Memories ZöhreKara Kayikci1), Dmitry Zaykovskiy1), Heiner Markert1 ... The mel filter block divides this spectrum into 23 equidistant, weighted, half-overlapping channels in the mel frequency domain ...
Research Interests:
Research Interests:
Measuring bio signals such as the heart rate in non medical applications is gaining an increasing importance. With camera based photoplethysmography (PPG) it is possible to measure the heart rate remotely with built in webcams of every... more
Measuring bio signals such as the heart rate in non medical applications is gaining an increasing importance. With camera based photoplethysmography (PPG) it is possible to measure the heart rate remotely with built in webcams of every tablet and laptop. Recent research with machine learning based methods showed great success compared to signal processing based methods. In this paper, we use k-nearest neighbor (kNN) and multilayer perceptron (MLP) with an alternative representation of the input vector. Estimating the quality of peaks with a Gaussian distribution could further improve the detection. Overall we could improve the root mean square error (RMSE) from 23.97 to 8.62.
Research Interests:
In this work, a method is presented for the continuous estimation of pain intensity based on fusion of bio-physiological and video features. The focus of the paper is to analyse which modalities and feature sets are suited best for the... more
In this work, a method is presented for the continuous estimation of pain intensity based on fusion of bio-physiological and video features. The focus of the paper is to analyse which modalities and feature sets are suited best for the task of recognizing pain levels in a person-independent setting. A large set of features is extracted from the available bio-physiological channels (ECG, EMG and skin conductivity) and the video stream. Experimental validation demonstrates which modalities contribute the most to a robust prediction and the effects when combining them to improve the continuous estimation given unseen persons.
Research Interests:
Research Interests:
Research Interests:
In this paper, a robust system for viewindependent action unit intensity estimation is presented. Based on the theory of sparse coding, region-specific dictionaries are trained to approximate the characteristic of the individual action... more
In this paper, a robust system for viewindependent action unit intensity estimation is presented. Based on the theory of sparse coding, region-specific dictionaries are trained to approximate the characteristic of the individual action units. The system incorporates landmark detection, face alignment and contrast normalization to handle a large variety of different scenes. Coupled with head pose estimation, an ensemble of large margin classifiers is used to detect the individual action units. The experimental validation shows that our system is robust against pose variations and able to outperform the challenge baseline by more than 35%.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
... 7 Distributed Architecture for Speech-Controlled Systems Based on Associative Memories ZöhreKara Kayikci1), Dmitry Zaykovskiy1), Heiner Markert1 ... The mel filter block divides this spectrum into 23 equidistant, weighted,... more
... 7 Distributed Architecture for Speech-Controlled Systems Based on Associative Memories ZöhreKara Kayikci1), Dmitry Zaykovskiy1), Heiner Markert1 ... The mel filter block divides this spectrum into 23 equidistant, weighted, half-overlapping channels in the mel frequency domain ...
Research Interests:
Research Interests:
This article joins the fields of emotion recognition and human computer interaction. While much work has been done on recognizing emotions, they are hardly used to improve a user’s interaction with a system. Although the fields of... more
This article joins the fields of emotion recognition and human computer interaction. While much work has been done on recognizing emotions, they are hardly used to improve a user’s interaction with a system. Although the fields of affective computing and especially serious games already make use of detected emotions, they tend to provide application and user specific adaptions only on the task level. We present an approach of utilizing recognized emotions to improve the interaction itself, independent of the underlying application at hand. Examining the state of the art in emotion recognition research and based on the architecture of Companion-System, a generic approach for determining the main cause of an emotion within the history of interactions is presented, allowing a specific reaction and adaption. Using such an approach could lead to systems that use emotions to improve not only the outcome of a task but the interaction itself in order to be truly individual and empathic.
Research Interests:
Die ventrikulare Spatpotentialanalyse im hochverstarkten EKG ist eine weitverbreitete nicht-invasive Methode, um Patienten mit einem erhohten Risiko ventrikularer Tachyar-rhythmien zu identifizieren und um eine Risikoabschatzung nach... more
Die ventrikulare Spatpotentialanalyse im hochverstarkten EKG ist eine weitverbreitete nicht-invasive Methode, um Patienten mit einem erhohten Risiko ventrikularer Tachyar-rhythmien zu identifizieren und um eine Risikoabschatzung nach Myokardinfarkt durchzufuhren. Bei der Spatpotentialanalyse im Zeitbereich werden ublicherweise drei Parameter extrahiert. Mit diesen Werten wird, anhand von empirisch ermittelten Grenzen, eine Aussage uber das Vorhandensein von Spatpotentialen, als niederamplitudigen, hoherfrequenten Signalen am Ende eines Herzschlages, gemacht [8,4].
Research Interests:
Bats are indicators for ecosystem health, and therefore the determination of bat activity and species abundance provides essential information for biodiversity research and conservation monitoring. In this study, we propose a... more
Bats are indicators for ecosystem health, and therefore the determination of bat activity and species abundance provides essential information for biodiversity research and conservation monitoring. In this study, we propose a computational method for the detection of bat echolocation calls. This method uses feature engineering and consists of a statistical model-based Voice Activity Detector combined with a Random Forests classifier (VAD+RF). Using an open-access library (www.batdetective.org), we trained and tested the performance of our method and compare it to other existing detection methods. These methods include a detector based on deep neural networks along with other commercial detection systems. To visualize the detector performance over the full range of possible class distributions and misclassification costs, we calculated the Cost Curves and \(F_1\)-measure Curves. Results show that the detecting power of VAD+RF is comparable to methods based on deep learning. Based on ...
Research Interests:
The focus of this work is emotion recognition in the wild based on a multitude of different audio, visual and meta features. For this, a method is proposed to optimize multi-modal fusion architectures based on evolutionary computing.... more
The focus of this work is emotion recognition in the wild based on a multitude of different audio, visual and meta features. For this, a method is proposed to optimize multi-modal fusion architectures based on evolutionary computing. Extensive uni- and multi-modal experiments show the discriminative power of each computed feature set and fusion architecture. Furthermore, we summarize the EmotiW 2013/2014 challenges and review the conclusions that have been drawn and compare our results with the state-of-the-art on this dataset.
Research Interests:
In this work, multi-modal fusion of video and biopotential signals is used to recognize pain in a person-independent scenario. For this purpose, participants were subjected to painful heat stimuli under controlled conditions.... more
In this work, multi-modal fusion of video and biopotential signals is used to recognize pain in a person-independent scenario. For this purpose, participants were subjected to painful heat stimuli under controlled conditions. Subsequently, a multitude of features have been extracted from the available modalities. Experimental validation suggests that the cues that allow the successful recognition of pain are highly similar across different people and complementary in the analysed modalities to an extent that fusion methods are able to achieve an improvement over single modalities. Different fusion approaches (early, late, trainable) are compared on a large set of state-of-the art features for the biopotentials and video channels in multiple classification experiments.
Research Interests:
A major building block for the construction of reliable statistical classifiers in the context of affective human-computer interaction is the collection of training samples that appropriately reflect the complex nature of the desired... more
A major building block for the construction of reliable statistical classifiers in the context of affective human-computer interaction is the collection of training samples that appropriately reflect the complex nature of the desired patterns. This is especially in this application a non-trivial issue as, even though it is easily agreeable that emotional patterns should be incorporated in future computer operating, it is by far not clear how it should be realized. There are still open questions such as which types of emotional patterns to consider together with their degree of helpfulness for computer interactions and the more fundamental question on what emotions do actually occur in this context. In this paper we start by reviewing existing corpora and the respective techniques for the generation of emotional contents and further try to motivate and establish approaches that enable to gather, identify and categorize patterns of human-computer interaction.
Research Interests:
Research Interests:
... IEEE Computer Press, 1996. 37. Bernhard Schölkopf , Christopher JC Burges , Alexander J. Smola, Advances in kernel methods: support vector learning, MIT Press, Cambridge, MA, 1999. 38. F. Schwenker. ... 2000. 39. F Schwenker and HA... more
... IEEE Computer Press, 1996. 37. Bernhard Schölkopf , Christopher JC Burges , Alexander J. Smola, Advances in kernel methods: support vector learning, MIT Press, Cambridge, MA, 1999. 38. F. Schwenker. ... 2000. 39. F Schwenker and HA Kestler. ...
Research Interests:
Die ventrikulare Spatpotentialanalyse im hochverstarkten EKG ist eine weitverbreitete nicht-invasive Methode, um Patienten mit einem erhohten Risiko ventrikularer Tachyar-rhythmien zu identifizieren und um eine Risikoabschatzung nach... more
Die ventrikulare Spatpotentialanalyse im hochverstarkten EKG ist eine weitverbreitete nicht-invasive Methode, um Patienten mit einem erhohten Risiko ventrikularer Tachyar-rhythmien zu identifizieren und um eine Risikoabschatzung nach Myokardinfarkt durchzufuhren. Bei der Spatpotentialanalyse im Zeitbereich werden ublicherweise drei Parameter extrahiert. Mit diesen Werten wird, anhand von empirisch ermittelten Grenzen, eine Aussage uber das Vorhandensein von Spatpotentialen, als niederamplitudigen, hoherfrequenten Signalen am Ende eines Herzschlages, gemacht [8,4].
Research Interests:
Research Interests:
Research Interests:
Gene interactions in cells can be represented by gene regulatory networks. A Boolean network models gene interactions according to rules where gene expression is represented by binary values (on / off or {1, 0}). In reality, however, the... more
Gene interactions in cells can be represented by gene regulatory networks. A Boolean network models gene interactions according to rules where gene expression is represented by binary values (on / off or {1, 0}). In reality, however, the gene's state can have multiple values due to biological properties. Furthermore, the noisy nature of the experimental design results in uncertainty about a state of the gene. Here we present a new Boolean network paradigm to allow intermediate values on the interval [0, 1]. As in the Boolean network, fixed points or attractors of such a model correspond to biological phenotypes or states. We use our new extension of the Boolean network paradigm to model gene expression in first and second heart field lineages which are cardiac progenitor cell populations involved in early vertebrate heart development. By this we are able to predict additional biological phenotypes that the Boolean model alone is not able to identify without utilizing additional ...
Research Interests:
Jede Art von Datenanalyse beginnt damit, das man sich zunachst einen Uberblick uber die vorliegenden Daten verschafft, wobei man vielleicht schon im Ansatz Strukturen erkennt. Dieses Vorstadium jedes noch so elaborierten... more
Jede Art von Datenanalyse beginnt damit, das man sich zunachst einen Uberblick uber die vorliegenden Daten verschafft, wobei man vielleicht schon im Ansatz Strukturen erkennt. Dieses Vorstadium jedes noch so elaborierten Datenanalysevebfahrens bleibt meist dem menschlichen Beobachter uberlassen. So verschafft uns beispielsweise bei einer grosen Menge von zweidimensionalen Datenpunkten ein optisches Display einen Uberblick. Es ist ein wichtiges Problem im Bereich der prarationalen maschinellen Intelligenz, diese scheinbar so einfache menschliche Fahigkeit nachzubilden. Allerdings ist auch der Mensch bei vielen hochdimensionalen Datenpunkten meist auserstande, sich einen Uberblick zu verschaffen. Hier konnten kluge Datenverarbeitungsmethoden helfen.
Research Interests:
Research Interests: Cognitive Science, Computer Science, Biomedical Engineering, Histology, Medicine, and 15 moreCats, Brain Mapping, Anatomy, Computers, Brain, Image Reconstruction, Mice, Animals, Surface, 3-D Reconstruction, Surface Area, Statistical Methods for Neuroscience, Cerebellar Cortex, Neurosciences, and Neuroscience Methods
The interactions between illusory and real contours have been investigated under monocular, binocular and dichoptic conditions. Results show that under all three presentation conditions, periodic alterations, generally called rivalry,... more
The interactions between illusory and real contours have been investigated under monocular, binocular and dichoptic conditions. Results show that under all three presentation conditions, periodic alterations, generally called rivalry, occur during the perception of cognitive (or illusory) triangles, while earlier research had failed to find such rivalry (Bradley and Dumais 1975). With line triangles, rivalry is experienced only under dichoptic conditions. A model is proposed to account for the observed phenomena.
Research Interests: Psychology, Cognitive Psychology, Cognitive Science, Mathematics, Computer Science, and 13 morePerception, Visual perception, Binocular vision, Medicine, Humans, Computer Simulation, Depth Perception, Binocular Rivalry, Optical Illusions, Rivalry, Biological cybernetics, Monocular Vision, and Illusory Contours
Despite of processing elements which are thousands of times faster than the neurons in the brain, modern computers still cannot match quite a few processing capabilities of the brain, many of which we even consider trivial (such as... more
Despite of processing elements which are thousands of times faster than the neurons in the brain, modern computers still cannot match quite a few processing capabilities of the brain, many of which we even consider trivial (such as recognizing faces or voices, or following a ...
Research Interests:
... 11, Monte carlo localization: Ecient position estimation for mobile robots - Fox, Burgard, et al. - 1999. 1, Visual feature detection for robot localization and object recognition in RoboCup," 2000, technical report -... more
... 11, Monte carlo localization: Ecient position estimation for mobile robots - Fox, Burgard, et al. - 1999. 1, Visual feature detection for robot localization and object recognition in RoboCup," 2000, technical report - Neubeck, Ritter, et al. ...
Research Interests:
... In Kohonen, T., editor, Artificial Neural Networks. ICANN-91, North-Holland. [Linde et al., 1980]Linde, Y., Buzo, A., and Gray, R.(1980). An algorithm for vector quantizer design. ... Berkeley University of California Press. [Moody... more
... In Kohonen, T., editor, Artificial Neural Networks. ICANN-91, North-Holland. [Linde et al., 1980]Linde, Y., Buzo, A., and Gray, R.(1980). An algorithm for vector quantizer design. ... Berkeley University of California Press. [Moody and Darken, 1989] Moody, J. and Darken, C.(1989). ...
Research Interests:
In Chap. 2 we defined thenovelty of a proposition as a special function of its probability p(A). We motivated the definition \(\mathcal{N}(A) = {-\log }_{2}p(A)\) by the idea that \(\mathcal{N}(A)\) should measure the number of yes–no... more
In Chap. 2 we defined thenovelty of a proposition as a special function of its probability p(A). We motivated the definition \(\mathcal{N}(A) = {-\log }_{2}p(A)\) by the idea that \(\mathcal{N}(A)\) should measure the number of yes–no questions needed to guess A.
Research Interests:
The set of all repertoires actually has an interesting structure, when we “look at” a repertoire α in terms of its proper descriptions D(α). This means that we should consider two repertoires to be essentially the same if they have the... more
The set of all repertoires actually has an interesting structure, when we “look at” a repertoire α in terms of its proper descriptions D(α). This means that we should consider two repertoires to be essentially the same if they have the same proper descriptions, or we should say that α is more refined than β if the proper descriptions in α are contained in those in β.
Research Interests:
Research Interests:
Research Interests:
This chapter introduces the notion of a cover or repertoire and its proper descriptions. Based on the new idea of relating covers and descriptions, some interesting properties of covers are defined.
Research Interests:
In this part we want to condense the new mathematical ideas and structures that have been introduced so far into a mathematical theory, which can be put in the framework of lattice theory. In the next chapter we want to get a better... more
In this part we want to condense the new mathematical ideas and structures that have been introduced so far into a mathematical theory, which can be put in the framework of lattice theory. In the next chapter we want to get a better understanding of the order structure (defined in Definition 10.3 on the set of all covers. For this purpose we now introduce a number of basic concepts concerning order and lattices (Birkhoff 1967).
Research Interests:
Research Interests:
This contribution presents the design of an image-based contextual pedestrian classifier for an automotive application. Our previous work shows that local classifiers working with image cutouts are in many cases not sufficient to achieve... more
This contribution presents the design of an image-based contextual pedestrian classifier for an automotive application. Our previous work shows that local classifiers working with image cutouts are in many cases not sufficient to achieve satisfactory results in complex scenarios. As a solution the work proposed incorporating contextual knowledge into the classification task, significantly improving the classification results. Contextual knowledge is described by a set of different and independent context sources. This paper discusses the fusion of these sources on the basis of the Dempster-Shafer theory. It presents and compares different possibilities to model the frame of discernment and the mass function to achieve optimal results. Furthermore, it provides an elegant way to take uncertainties of the context sources into account. The methods are evaluated on simulated and on real data.
Research Interests:
The approximation of kernel functions using explicit feature maps gained a lot of attention in recent years due to the tremendous speed up in training and learning time of kernel-based algorithms, making them applicable to very... more
The approximation of kernel functions using explicit feature maps gained a lot of attention in recent years due to the tremendous speed up in training and learning time of kernel-based algorithms, making them applicable to very large-scale problems. For example, approximations based on random Fourier features are an efficient way to create feature maps for a certain class of scale invariant kernel functions. However, there are still many kernels for which there exists no algorithm to derive such maps. In this work we propose an efficient method to create approximate feature maps from an arbitrary distance metric using pseudo line projections called Distance-Based Feature Map (DBFM). We show that our approximation does not depend on the input dataset size or the dimension of the input space. We experimentally evaluate our approach on two real datasets using two metric and one non-metric distance function.
Research Interests:
Research Interests:
In most behavior-based approaches, implementing a broad set of different behavioral skills and coordinating them to achieve coherent complex behavior is an error-prone and very tedious task. Concepts for organizing reactive behavior in a... more
In most behavior-based approaches, implementing a broad set of different behavioral skills and coordinating them to achieve coherent complex behavior is an error-prone and very tedious task. Concepts for organizing reactive behavior in a hierarchical manner are rarely found in behavior-based approaches, and there is no widely accepted approach for creating such behavior hierarchies. Most applications of behavior-based concepts use
Research Interests:
After the work of Marmarelis & Naka (1972, 1973) in the catfish retina, systems analysis using stochastic stimuli has had a boom in the seventies (e.g. McCann & Marmarelis, 1975; Eckert & Bishop, 1975; French & Wong, 1977; Lipson, 1975;... more
After the work of Marmarelis & Naka (1972, 1973) in the catfish retina, systems analysis using stochastic stimuli has had a boom in the seventies (e.g. McCann & Marmarelis, 1975; Eckert & Bishop, 1975; French & Wong, 1977; Lipson, 1975; McCann, 1974; Naka, Marmarelis & Chan, 1975; Spekreijse & Reits, 1982; Trimble & Phillips, 1978; Terzuolo et al. 1982). White-noise analysis was considered to be a general tool for investigating nonlinear systems gaining a maximum of information with a minimum of assumptions about the system. The modification of the original Wiener theory (Wiener, 1958; Cameron & Martin, 1947; McKean, 1972) by Lee & Schetzen (1965) made the theory fairly easy to implement into widely available computers and thus accessible to a larger number of experimenters.