Faster R-CNN and YOLO based Vehicle detection: A Survey

Automatic moving vehicle detection plays a crucial and challenging role in performing intelligent traffic surveillance. Numerous research projects aiming to perform proper detection and tracking of vehicles have been carried out and the methods designed under these projects have found their uses in various important applications for e.g. to minimize the fatal accidents which mainly occur due to negligence of drivers or due to poor visibility during inclement weather condition or due to improper illumination, etc. At present, several deep neural networks have been proposed for performing object detection. This paper presents a comprehensive review of existing Faster Region-based Convolutional Neural Network (Faster R-CNN) and You look only once (YOLO) based vehicle detection and tracking methods. In this survey, we have divided the existing vehicle detection methods into different groups depending upon the architecture (Faster R-CNN/YOLO) which have been used as the backbone of these designed methods. We have organized the entire survey in chronological order so that interrelations between proposed methods can be highlighted. Apart from performing in depth analyses of the existing methods, we have described the respective architectures of Faster R-CNN, YOLO and their proposed variants in details in this survey for better understanding. We have concluded this paper by listing down the limitations of the existing works and unexplored aspects of this research topic. We have also thrown some light on the future scope of this research area.