کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
431523 | 688565 | 2012 | 12 صفحه PDF | دانلود رایگان |
Self-organising neural models have the ability to provide a good representation of the input space. In particular the Growing Neural Gas (GNG) is a suitable model because of its flexibility, rapid adaptation and excellent quality of representation. However, this type of learning is time-consuming, especially for high-dimensional input data. Since real applications often work under time constraints, it is necessary to adapt the learning process in order to complete it in a predefined time. This paper proposes a Graphics Processing Unit (GPU) parallel implementation of the GNG with Compute Unified Device Architecture (CUDA). In contrast to existing algorithms, the proposed GPU implementation allows the acceleration of the learning process keeping a good quality of representation. Comparative experiments using iterative, parallel and hybrid implementations are carried out to demonstrate the effectiveness of CUDA implementation. The results show that GNG learning with the proposed implementation achieves a speed-up of 6×6× compared with the single-threaded CPU implementation. GPU implementation has also been applied to a real application with time constraints: acceleration of 3D scene reconstruction for egomotion, in order to validate the proposal.
► We propose the implementation of the GNG on a GPGPU architecture.
► We present a detailed study to obtain the best performance with CUDA.
► The GPGPU implementation is able to deal with time restriction
► Acceleration of the 3D scene reconstruction with GNG onto GPUs with CUDA.
► We test our work performing six degrees of freedom (6DoF) pose registration.
Journal: Journal of Parallel and Distributed Computing - Volume 72, Issue 10, October 2012, Pages 1361–1372