A survey of deep learning methods for density estimation and crowd counting

Abstract Accurately estimating the number of objects in a single image is a challenging yet meaningful task and has been applied in many applications such as urban planning and public safety. In various object counting tasks, crowd counting is particularly prominent due to its specific significance to social security and development. Fortunately, the development of the techniques for crowd counting can be generalized to other related fields such as vehicle counting and environment survey, if without taking their characteristics into account. Therefore, many researchers are devoting to crowd counting, and many excellent works have spurted out and significantly promote the development of crowd counting. However, one question we should consider is how far are we from solving the counting problem?. Limited by the costs of time and energy, we cannot analyze all algorithms. In this paper, we have surveyed 300+ works to comprehensively and systematically study the crowd counting models, mainly involving CNN-based density map estimation methods. Finally, according to evaluation metrics, we select the top three performers on their crowd counting datasets and analyze their merits and drawbacks. Through our analysis, we expect to make a reasonable inference and prediction for the future development of crowd counting, and meanwhile, it can also provide feasible solutions for the problem of object counting in other fields. We provide density maps and prediction results of some mainstream algorithms on the validation set of NWPU dataset for comparison and testing. Meanwhile, density map generation and evaluation tools are also provided. All the codes and evaluation results are made publicly available at https://github.com/gaoguangshuai/survey-for-crowd-counting .