Object detection and recognition is an important and challenging problem of computer vision. With the evolution in the field of Machine Learning, many researchers and practitioners extensively experimented and enhanced the object detection technique performance. Object detection and recognition have found widespread application in recent years, particularly in intelligent transport systems, healthcare, and character recognition. Generally, object detection technique performance is examined by its inference time and accuracy. We begin by examining the fundamental principles of object detection. Subsequently, this manuscript provides a detailed analysis of the single-stage You Look Only Once (YOLO) model and its subsequent versions which examine feature extraction methods, mosaic transformation and compound scaling. We have also presented detailed comparison table of different YOLO versions.