The practical implementations and technological advances of the YOLO method in computer vision for object detection are examined in this article. Santosh Divvala and Joseph Redmon’s YOLO artificial intelligence system can accurately and quickly detect objects in videos and images, eliminating the need for network iterations. Instead of using traditional methods, YOLO directly generates bounding boxes and class probabilities for each object, optimising the procedure. After reading this, the process is clearer. This study examines YOLO’s effectiveness in various situations. This technology has many applications in industrial automation, AR/VR, autonomous vehicles, healthcare, and retail. This article examines YOLO model training and data preparation. COCO datasets are used in this study. The architectural components of YOLO versions 7 and 8 are also examined. The findings show that YOLO v8 is highly accurate in identifying a wide range of objects in real time. This study shows YOLO’s revolutionary potential as object detection software for BV applications, highlighting its outstanding features.