A comprehensive survey of the R-CNN family for object detection

Object detection using deep learning, one of the most challenging problems in computer vision, seeks to locate instances of objects from a large number of predefined categories in natural images. Given this period of rapid evolution, the main contribution of this paper is to provide a comprehensive survey of the region-based convolutional neural network (R-CNN) family (R-CNN, Fast R-CNN, and Faster R-CNN). In comparison to the R-CNN and Fast R-CNN, simulation results show that the faster R-CNN improves not only accuracy but also detection speed. For robust object detection, it has been found that the Faster R-CNN is particularly suited for this purpose. We conclude with several open issues and challenges that are keys to the design of future work.