Deep learning has become very prominent on how the world has adapted to Artificial Intelligence in recent years. Object tracking is an application of deep learning in which the programme takes a series of initial object detections and creates a unique identifier or ids for each of them, then tracks the detected objects as they move around frames in a video. Faster-RCNN and SSD have superior accuracy among the algorithms, whereas YOLO performs great when speed is prioritized over accuracy. For security purpose or for any important tasks to be done, a very efficient method is needed. So that it minimizes the errors and improves the performance that leads to great applications. To achieve all the desired factors, in this paper we are using YOLO (You Only Look Once) and Deep SORT from deep learning to detect and track the objects. We employ python as a programming language and the latest version of YOLO to achieve successful outcomes by merging these two. When all of these factors are combined, the outcome is excellent for object detection and tracking, as well as for calculating other parameters such as object counting.

Object detection, YOLO, Deep learning, Vehicle tracking, Deep SORT