This book on "Real Time Background Subtraction using CUDA Architecture" is written with a viewpoint of giving an exposure to the readers about complete process of Background subtraction (BGS) which is one of the important and crucial step of many computer vision applications such as airport shopping complexes, traffic monitoring system etc. where we want to segment out the foreground objects from the background of a video. Chapter 1 outlines the introduction of BGS and GPU along with technical challenges to develop a good background subtraction system. Chapter 2 describes detailed literature survey of existing systems whereas Chapter 3 presents different background subtraction techniques with their advantages and limitations. Chapter 4 highlights the technique of Thresholding and data validation step of foreground extraction process. Chapter 5 discusses implementation of system using GPU. It outlines the data flow diagrams of the system and the flow of data using CUDA architecture for parallel processing. In Chapter 6 we discussed various techniques of CUDA optimization. In Chapter 7, we demonstrated experimental results and its analysis based on performance evaluation parameters.