Refine
Year of publication
- 2015 (11) (remove)
Document Type
- Conference Proceeding (10)
- Article (1)
Language
- English (11) (remove)
Has Fulltext
- no (11)
Is part of the Bibliography
- no (11)
We present a novel approach of distributing matrix multiplications among GPU-equipped nodes in a cluster system. In this context we discuss the induced challenges and possible solutions. Additionally we state an algorithm which outperforms optimized GPU BLAS libraries for small matrices. Furthermore we provide a novel theoretical model for distributing algorithms within homogeneous computation systems with multiple hierarchies. In the context of this model we develop an algorithm which can find the optimal distribution parameters for each involved subalgorithm. We provide a detailed analysis of the algorithms space and time complexities and justify its use with a structured evaluation within a small GPU-equipped Beowulf cluster.
We present a novel method to perform multi-class pattern classification with neural networks and test it on a challenging 3D hand gesture recognition problem. Our method consists of a standard one-against-all (OAA) classification, followed by another network layer classifying the resulting class scores, possibly augmented by the original raw input vector. This allows the network to disambiguate hard-to-separate classes as the distribution of class scores carries considerable information as well, and is in fact often used for assessing the confidence of a decision. We show that by this approach we are able to significantly boost our results, overall as well as for particular difficult cases, on the hard 10-class gesture classification task.
A light-weight real-time ap- plicable hand gesture recognition system for automotive applications
(2015)
We present a novel approach for improved hand-gesture recognition by a single time-of-flight(ToF) sensor in an automotive environment. As the sensor's lateral resolution is comparatively low, we employ a learning approach comprising multiple processing steps, including PCA-based cropping, the computation of robust point cloud descriptors and training of a Multilayer perceptron (MLP) on a large database of samples. A sophisticated temporal fusion technique boosts the overall robustness of recognition by taking into account data coming from previous classification steps. Overall results are very satisfactory when evaluated on a large benchmark set of ten different hand poses, especially when it comes to generalization on previously unknown persons.
We present a system for efficient dynamic hand gesture recognition based on a single time-of-flight sensor. As opposed to other approaches, we simply rely on depth data to interpret user movement with the hand in mid-air. We set up a large database to train multilayer perceptrons (MLPs) which are subsequently used for classification of static hand poses that define the targeted dynamic gestures. In order to remain robust against noise and to balance the low sensor resolution, PCA is used for data cropping and highly descriptive features, obtainable in real-time, are presented. Our simple yet efficient definition of a dynamic hand gesture shows how strong results are achievable in an automotive environment allowing for interesting and sophisticated applications to be realized.
We present a novel hierarchical approach to multi-class classification which is generic in that it can be applied to different classification models (e.g., support vector machines, perceptrons), and makes no explicit assumptions about the probabilistic structure of the problem as it is usually done in multi-class classification. By adding a cascade of additional classifiers, each of which receives the previous classifier's output in addition to regular input data, the approach harnesses unused information that manifests itself in the form of, e.g., correlations between predicted classes. Using multilayer perceptrons as a classification model, we demonstrate the validity of this approach by testing it on a complex ten-class 3D gesture recognition task.
Utilizing biometrie traits for privacy- and security-applications is receiving an increasing attention. Applications such as personal identification, access control, forensics appli-cations, e-banking, e-government, e-health and recently person-alized human-smart-home and human-robot interaction present some examples. In order to offer person-specific services for/of specific person a pre-identifying step should be done in the run-up. Using biometric in such application is encountered by diverse challenges. First, using one trait and excluding the others depends on the application aimed to. Some applications demand directly touch to biometric sensors, while others don't. Second challenge is the reliability of used biometric arrangement. Civilized application demands lower reliability comparing to the forensics ones. And third, for biometric system could only one trait be used (uni-modal systems) or multiple traits (Bi- or Multi-modal systems). The latter is applied, when systems with a relative high reliability are expected. The main aim of this paper is providing a comprehensive view about biometric and its application. The above mentioned challenges will be analyzed deeply. The suitability of each biometric sensor according to the aimed application will be deeply discussed. Detailed com-parison between uni-modal and Multi-modal biometric system will present which system where to be utilized. Privacy and security issues of biometric systems will be discussed too. Three scenarios of biometric application in home-environment, human-robot-interaction and e-health will be presented.
As smart homes are being more and more popular, the needs of finding assisting systems which interface between users and home environments are growing. Furthermore, for people living in such homes, elderly and disabled people in particular and others in general, it is totally important to develop devices, which can support and aid them in their ordinary daily life. We focused in this work on sustaining privacy issues of the user during a real interaction with the surrounding home environment. A smart person-specific assistant system for services in home environment is proposed. The role of this system is the assisting of persons by controlling home activities and guiding the adaption of Smart-Home-Human interface towards the needs of the considered person. At the same time the system sustains privacy issues of it’s interaction partner. As a special case of medical assisting the system is so implemented, that it provides for elderly or disabled people person-specific medical assistance . The system has the ability of identifying its interaction partner using some biometric features. According to the recognized ID the system, first, adopts towards the needs of recognized person. Second the system represents person-specific list of medicines either visually or auditive. And third the system gives an alarm in the case of taking medicament either later or earlier as normal taking time.
We present a novel approach of distributing small-to mid-scale neural networks onto modern parallel architectures. In this context we discuss the induced challenges and possible solutions. We provide a detailed theoretical analysis with respect to space and time complexities and reinforce our computation model with evaluations which show a performance gain over state of the art approaches.
Object detection systems which operate on large data streams require an efficient scaling with available computation power. We analyze how the use of tile-images can increase the efficiency (i.e. execution speed) of distributed HOG-based object detectors. Furthermore we discuss the challenges of using our developed algorithms in practical large scale scenarios. We show with a structured evaluation that our approach can provide a speed-up of 30-180 % for existing architectures. Due to the its generic formulation it can be applied to a wide range of HOG-based (or similar) algorithms. In this context we also study the effects of applying our method to an existing detector and discuss a scalable strategy for distributing the computation among nodes in a cluster system.
This contribution demonstrates the efficient embedding of a single depth-camera into the automotive environment making mid-air gesture interaction for mobile applications viable in such a scenario. In this setting a new human-machine interface is implemented to give an idea of future improvements in automation processes in industrial applications. Our system is based on a data-driven approach by learning hand poses as well as gestures from a large database in order to apply them on mobile devices. We register any movement in a nearby driver area and crop data efficiently with the means of PCA transforming it into so-called feature vectors which present the input for our multi-layer perceptrons (MLPs). After MLP classification, the interpretation of user input is sent via WiFi to a tablet PC mounted into the car interior visualizing an infotainment system which the user is able to interact with. We demonstrate that by this setup hand gestures as well as hand poses are easily and efficiently interpretable insofar as that they become an intuitive and supplementary means of interaction for automotive HMI in mobile scenarios realizable in real-time.