CUDA implementation of the pixel based adaptive segmentation algorithm

Author

Karahan, Samil ; Sevilgen, Fatih Erdogan

Author_Institution

Bilisim Teknolojileri Enstitusu, Tubitak Bilgem, Kocaeli, Turkey

fYear

2015

fDate

16-19 May 2015

Firstpage

2505

Lastpage

2508

Abstract

The operation of foreground background subtraction, which has a crucial role on video processing, needs to be processed in real time. In this paper, we present an implement of the pixel based adaptive segmentation (PBAS) algorithm on CUDA parallel programming platform to process high resolution videos in real time. The performance of the implementation is enhanced by using Nsight profiler metrics such as consumed time for functions, usage of GPU compute and memory, occupancy, divergence. Experimental evaluation on benchmark videos presents the performance of the implementation; On a Tesla K20 graphic processing unit, 646 fps is achieved for a video with 320×240 resolution when memory transfer to and from the GPU is included. As a result, about 17 time speedup is achieved with regards to single thread version.

Keywords

graphics processing units; image resolution; image segmentation; parallel architectures; parallel programming; video signal processing; CUDA implementation; CUDA parallel programming platform; GPU; Nsight profiler metrics; PBAS algorithm; Tesla K20 graphic processing unit; foreground background subtraction; high resolution video processing; memory transfer; pixel based adaptive segmentation algorithm; real time; single thread version; Art; Benchmark testing; Computer vision; Conferences; Graphics processing units; Real-time systems; Streaming media; Adaptif Foreground Background Substraction; CUDA; Nsight Profiler;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing and Communications Applications Conference (SIU), 2015 23th

Conference_Location

Malatya

Type

conf

DOI

10.1109/SIU.2015.7130393

Filename

7130393