A Survey of Deep Learning-Based Object Detection

Abstract

Object detection is one of the most important and challenging branches of\ncomputer vision, which has been widely applied in peoples life, such as\nmonitoring security, autonomous driving and so on, with the purpose of locating\ninstances of semantic objects of a certain class. With the rapid development of\ndeep learning networks for detection tasks, the performance of object detectors\nhas been greatly improved. In order to understand the main development status\nof object detection pipeline, thoroughly and deeply, in this survey, we first\nanalyze the methods of existing typical detection models and describe the\nbenchmark datasets. Afterwards and primarily, we provide a comprehensive\noverview of a variety of object detection methods in a systematic manner,\ncovering the one-stage and two-stage detectors. Moreover, we list the\ntraditional and new applications. Some representative branches of object\ndetection are analyzed as well. Finally, we discuss the architecture of\nexploiting these object detection methods to build an effective and efficient\nsystem and point out a set of development trends to better follow the\nstate-of-the-art algorithms and further research.\n

Keywords

Affiliated Institutions

• Xidian University CN
• University of Nottingham GB

Related Publications