期刊名称：International Journal of Image Processing (IJIP)
电子版ISSN：1985-2304
出版年度：2012
卷号：6
期号：6
页码：441-466
出版社：Computer Science Journals
摘要：Object detection and recognition are important problems in computer vision. Since these problems are meta-heuristic, despite a lot of research, practically usable, intelligent, real-time, and dynamic object detection/recognition methods are still unavailable. The accuracy level of any algorithm or even Google glass project is below 16% for over 22,000 object categories. With this accuracy, it's practically unusable. This paper reviews the various aspects of object detection and the challenges involved. The aspects addressed are feature types, learning model, object templates, matching schemes, and boosting methods. Most current research works are highlighted and discussed. Decision making tips are included with extensive discussion of the merits and demerits of each scheme.
