Refine
Document Type
- Conference Proceeding (11)
- Article (1)
- Report (1)
Language
- English (13) (remove)
Has Fulltext
- no (13)
Is part of the Bibliography
- no (13)
In this paper, we describe an efficient method for a fast people re-identification based on models of human clothes. An initial model is estimated during people detection and tracking, which will be refined during the re-identification. This stepwise extraction, combination and comparing of features speeds up the whole re-identification. For the refining, several saliency maps are used to extract individual features. These individual features are located separately for any human body part. The body parts are located with an optimized GPU-based HOG detector. Furthermore, we introduce a meanshift-based fusion concept which utilizes multiple detectors in order to increase the detection reliability.
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.
In the context of existing approaches to cluster computing we present a newly developed modular framework `SimpleHydra' for rapid deployment and management of Beowulf clusters. Instead of focusing only the pure computation tasks on homogeneous clusters (i.e. clusters with identically set up nodes), this framework aims to ease the configuration of heterogeneous clusters and to provide a low-level / high-level object-oriented API for low-latency distributed computing. Our framework does not make any restrictions regarding the hardware and minimizes the use of external libraries to the case of special modules. In addition to that our framework enables the user to develop highly dynamic cluster topologies. We describe the framework's general structure as well as time critical elements, give application examples in the `Big-Data' context during a research project and briefly discuss additional features. Furthermore we give a thorough theoretical time/space complexity analysis of our implemented methods and general approaches.
In this paper, we describe a method to model human clothes for a later recognition by the use of RGB- and SWIR-cameras. A basic model is estimated during people detection and tracking. This model will be refined if the recognition is triggered. For the refining, several saliency maps are used to extract individual features. These individual features are located separately for any human body parts. The body parts are estimated by the use of a silhouette extraction combined with a skeleton estimation. In this way, the model describes the human clothes in a compact manner which allows the use of a simple and fast comparison method for people recognition. Such models can be used in security and service applications.
Autonomous robots with limited computational capacity call for control approaches that generate meaningful, goal-directed behavior without using a large amount of resources. The attractor dynamics approach to movement generation is a framework that links sensor data to motor commands via coupled dynamical systems that have attractors at behaviorally desired states. The low computational demands leave enough system resources for higher level function like forming a sequence of local goals to reach a distant one. The comparatively high performance of local behavior generation allows the global planning to be relatively simple. In the present paper, we apply this approach to generate walking trajectories for a small humanoid robot, the Aldebaran Nao, that are goal-directed and avoid obstacles. The sensor information is a single camera in the head of the robot. The limited field of vision is compensated by head movements. The design of the dynamical system for motion generation and the choice of state variable makes a computationally expensive scene representation or local map building unnecessary.
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.
Technical Report
(2016)
This internal report discusses the theoretical and practical aspects of the cluster management framework SimpleHydra, which was developed in order to allow researchers the quick setup of classical small to mid-scale computation clusters while being as lightweight and platform independent as possible. We motivate crucial design choices with a theoretical analysis in the aspect of time and space complexity, furthermore we give a comprehensive introduction regarding the frameworks usage (which includes examples and detailed description of fundamental concepts as well as data structures). In addition to that we illustrate application scenarios with complete source code examples. Furthermore we hope that this document proves valuable not only as a development report but also as a practical manual for SimpleHydra.