high performance computing on graphics processing units: hgpu.org

Feature vectors can be anything from simple surface normals to more complex feature descriptors. Feature extraction is important in order to solve various computer vision problems: e.g. registration, object recognition and scene understanding. Most of these techniques cannot be computed online due to their complexity and the context where they are applied. Therefore computing these […]

Virtual surgery simulators are emerging as a training method for medical specialists and are expected to provide a virtual environment that is realistic and responsive enough to be able to physically simulate a wide variety of medical scenarios. Haptic interaction with the environment requires an underlying physical model that is dynamic, deformable, and computable in […]

Simulating light is a very computationally expensive proposition. There are a wide variety of global illumination algorithms that are implemented and used by major motion picture companies to render interesting and believable scenes. Every algorithm strives to find a balance between speed and accuracy. The Point Based Approximate Color Bleeding algorithm is one of the […]

CUDA is an advanced massively parallel computing platform that can provide high performance computing power at much more affordable cost. In this paper, we present a parallel graph-based substructure pattern mining algorithm using CUDA Dynamic Parallelism. The key contribution is a parallel solution to traversing the DFS (Depth First Search) code tree. Furthermore, we implement […]

Filtered back-projection algorithms are widely used for the reconstruction of volumetric data from cone-beam projections in interventional C-arm computed tomography. Furthermore, general-purpose GPUs have become a popular tool for accelerating the reconstruction during time-critical clinical procedures. In this work, we focus on the systematic performance optimization of cone-beam back-projection on the latest architecture of CUDA-enabled […]

Discrete Cosine Transform (DCT) is a technique to get frequency separation. When DCT is applied on an image, it will give frequency segregation of an image since it is composed of DC value and range of low frequency values to high frequency values. DCT is very useful in image compression. When high frequency values are […]

This Thesis intends to be a startup guide in understanding the basics of Image Processing techniques and common use cases, but at the same time take advantage of the Graphics Processing Unit available in today’s embedded multimedia microprocessors present in netbooks, smartphones and tablets.

Mobility is a key element of many processes and activities, and the understanding of movement is important in many areas of science and technology. With the recent advances in technologies for mobile devices, like GPS and mobile phones, we are able to generate data sets of people, animals, vehicles and other moving objects, normally available […]

Sparse matrix-vector multiplication (spMVM) is the most time-consuming kernel in many numerical algorithms and has been studied extensively on all modern processor and accelerator architectures. However, the optimal sparse matrix data storage format is highly hardware-specific, which could become an obstacle when using heterogeneous systems. Also, it is as yet unclear how the wide single […]

Cell nuclei extraction from histopathology images is necessary for breast cancer grading, and has become one of the major problem in the domain of automatic image analysis. Stochastic marked point processes combined with birth and death processes are promising tools for such extraction, but they are extremely compute intensive, especially on large images such as […]

In recent years, Artificial Neural networks (ANNs) have been intensively employed to build smart model of microwave devices. In this paper a characterization of lossy SIW resonators by means of Multilayer Perceptron Neural Networks (MLPNNs) on Graphics Processing Unit (GPU), is presented. Once properly selected and trained, a MLPNN can evaluate the lossy SIW resonator’s […]

This paper presents a new method for representing Light Propagation Volumes using an octree data structure, and for allowing light from regular point light sources to be injected into them. The resulting technique uses full octrees with the help of a separate data structure for representing the octree structure. The octree structure enables light propagation […]