부산대학교 도서관

Several methods have been put out in the past few years in the field of computer vision for the goal of obj ect detection. The offered methods carry out the task of classifying and locating an object inside an image or a series of photos. Modern object detection frameworks and techniques are referred to as evolutionary approaches. The challenges of obj ect detection in numerous photos and movies are addressed through evolutionary techniques. Using the spatial information of the object in the image frames, the evolutionary approach based approaches may detect objects. By sliding a window across the object instances in the given image sequence, these algorithms can extract the features of the objects. In this paper, we have reviewed the object detection field's evolutionary strategy. The backend CNN architectures, CNN-based object identification techniques, and common object detection databases are discussed in this paper. This paper primarily focus on dataset detection and finding suitable object detection method. We have also provided a list of open source frameworks for academics and students who may be interested in using them to create object identification methods and algorithms.