GPU based background subtraction using CUDA: State of the Art

Fundamental to advance video processing such as object tracking, gait recognition, and video indexing is the issue of robust background modeling. Regarding these applications, several methods have been explored, but they are having high time complexity. The processing of real time video is computationally intensive. In recent years, many application developers are attracted towards GPUs as it is a powerful massively parallel system. Since the GPUs provide a vast number of multithreaded cores which are simple and data-parallel in nature with high memory bandwidths, they have been used to provide a valid support in the video processing research area. CUDA as a general purpose parallel computing architecture make GPUs an appealing choice to solve many complex computational problems in a more efficient way. In this paper, the background subtraction techniques and relevant applications are reviewed, highlighting the advantages of using CUDA in terms of efficiency.