No abstract

We introduce a GPU grid-based data structure for massively parallel nearest neighbor searches for dynamic point clouds. The implementation provides real-time performance and it is executed on GPU, both grid construction and nearest neighbors (approximate or exact) searches. This minimizes the memory transfer between device and system memories, improving overall performance. The proposed algorithm may be used across different applications with static and dynamic scenarios. Moreover, our data structure supports three-dimensional point clouds and given its dynamic nature, the user can change the data structure's parameters at runtime. The same applies to the number of neighbors to be found. Performance comparisons were made against previous works, endorsing the benefits of our solution. Finally, we were able to develop a real-time Point-Based Rendering application for validation of the data structure. Its drawbacks and data distribution's impact on performance were analysed and some directions for further investigation are given.

The main purpose of this survey is presenting the potential of GPGPU technology for real time markerless augmented reality related processing. CUDA is a GPGPU technology developed by NVIDIA that allows programmers to use the C programming language to code algorithms for execution on the GPU. Applications that require mathematically intensive computation of large amounts of data are ideal targets for GPU computing. In this survey, CUDA architecture will be depicted, together with an optimized programming model for obtaining better results using the parallel approach. A case study, mainly related to tracking algorithms, will also be shown in order to demonstrate the performance improvement in comparison to sequential approaches.