Given the massive increase of data, also arises the need for a system that efficiently handles it. Moving object tracking in a video is a crucial job in computer vision. This can be defined as the problem of estimating the trajectories of the object of interest in the video sequence. Tracking of objects is quite easy for human being as they have good vision and perception capability. However, for a machine, it is a task which is hard to achieve. Moving objects tracking has been an active research area for the last few decades. So many works have been done, but still it is an open problem which needs to be addressed for a good and robust object tracking system. Researchers, computer scientists, and academics have been working hard over the last several decades to solve the problems of object detection issue. In essence, object detection is used to identify objects in images and videos. There has been a lot of work done in this area at the start of the twenty-first century, which is performing well but cannot be effectively employed for real-time detection due to their slow data processing speed and poor precision. Additionally, Convolution neural networks underwent a quick transformation throughout the deep learning period, opening up new research directions. To date, much good work has been done in this area, including the region-based convolution network (R-CNNs), YOLO, etc. This paper reviews various components of object identification and tracking.