Efficient adaptive load balancing approach for compressive background subtraction algorithm on heterogeneous CPU–GPU platforms

• Published in: Journal of real-time image processing 2020-10, Vol.17 (5), p.1567-1583
• Main Authors: Mabrouk, Lhoussein, Huet, Sylvain, Houzet, Dominique, Belkouch, Said, Hamzaoui, Abdelkrim, Zennayi, Yahya
• Format: Article
• Language: English
• Subjects: Accuracy; Algorithms; Central processing units; Computer Graphics; Computer Science; CPUs; Distributed, Parallel, and Cluster Computing; Embedded Systems; Energy efficiency; Fourier transforms; Graphics processing units; Hardware Architecture; Image Processing and Computer Vision; Microprocessors; Multimedia Information Systems; Original Research Paper; Partitioning; Pattern Recognition; Platforms; Processors; Signal,Image and Speech Processing; Workloads
• ISSN: 1861-8200; 1861-8219
• DOI: 10.1007/s11554-019-00916-4

Mixture of Gaussians (MoG) and compressive sensing (CS) are two common approaches in many image and audio processing systems. The combination of these algorithms is recently used for the compressive background subtraction task. Nevertheless, the result of this combination has not been exploited to take advantage of the evolution of parallel computing architectures. This paper proposes an efficient strategy to implement CS-MoG on heterogeneous CPU–GPU computing platforms. This is achieved through two elements. The first one is ensuring the better acceleration and accuracy that can be achieved for this algorithm on both CPU and GPU processors: The obtained results of the improved CS-MoG are more accurate and performant than other published MoG implementations. The second contribution is the proposition of the Optimal Data Distribution Cursor ODDC, a novel adaptive data partitioning approach to exploit simultaneously the heterogeneous processors on any given platform. It aims to ensure an automatic workload balancing by estimating the optimal data chunk size that must be assigned to each processor, with taking into consideration its computing capacity. Furthermore, our method ensures an update of the partitioning at runtime to take into account any influence of data content irregularity. The experimental results, on different platforms and data sets, show that the combination of these two contributions allows reaching 98% of the maximal possible performance of targeted platforms.