A Review: Object Detection Algorithms

Object identification continues to be one of the most complex applications in computer vision and picture comprehension. Traditional machine learning object recognition prioritizes the extraction of interpretable properties higher. Deep neural network modelling and better object representation have greatly increased the precision of object detection. This study presents a closer look at how object identification has evolved in the deep learning era over time. Our paper will concentrate on techniques that are currently used in the community and how each of them varies from others. First, a quick summary of object detection as both tactic and tool shall be given. This will be followed by a quick discussion of essential steps and the organisation of object detection techniques. These methods include single-stage models like the YOLO series from YOLOv1 to YOLOv7, SSD and RetinaNet along with some two-stage models. Finally, we will wrap things off by outlining some potential analyses like comparison, applications, and drawbacks in that area, to suggest various strategies that might be used in various industries to get the best outcomes.