Abstract
Parking issues are becoming more significant as the number of drivers in crowded urban centers increases. At the same time, new breakthroughs in the fields of internet-of-things, computer vision and machine learning have opened up new opportunities for maximizing parking efficiency. In this paper, we present a comparative study of the existing approaches that use convolutional neural networks based smart parking solutions. We will explore how well various experiments using convolutional neural networks have dealt with the complex problems surrounding parking space identification, such as occlusions, different lighting conditions, and difference in car orientations.