Review on Deep based Object Detection

摘要

Object detection aims to detect and recognize all the salient targets in the whole image, which is one of the most fundamental and significant problems in computer vision. With the rapid development of deep learning-based detection algorithms, the performance of object detectors has been greatly improved. Thus, based on this period of rapid development, the purpose of this paper is to provide a brief survey of the latest achievements and gives people a quick overview of the latest achievements in this field brought about by deep learning techniques. In this survey, deep based object detection is categorized, covering some well-known one-stage and two-stage detectors. Moreover, the mainstream object detection datasets are listed, and the evaluation metrics are also provided for them. A novel branch of the object detection dataset (MaSTr1325) is analyzed as well. This survey also gives an in-depth perspective on future research.