A Strong Baseline for Crowd Counting and Unsupervised People Localization
This work provides an improved baseline and an unsupervised localization method for researchers and practitioners working on crowd counting, offering strong specific gains.
This paper establishes a strong baseline for crowd counting by evaluating various backbones and training tricks, achieving significant reductions in MAE and RMSE across multiple datasets. Additionally, it introduces an unsupervised clustering algorithm called isolated KMeans for localizing heads in density maps.
In this paper, we explore a strong baseline for crowd counting and an unsupervised people localization algorithm based on estimated density maps. Firstly, existing methods achieve state-of-the-art performance based on different backbones and kinds of training tricks. We collect different backbones and training tricks and evaluate the impact of changing them and develop an efficient pipeline for crowd counting, which decreases MAE and RMSE significantly on multiple datasets. We also propose a clustering algorithm named isolated KMeans to locate the heads in density maps. This method can divide the density maps into subregions and find the centers under local count constraints without training any parameter and can be integrated with existing methods easily.