Reducing Spatial Labeling Redundancy for Active Semi-Supervised Crowd Counting

Labeling is onerous for crowd counting as it should annotate each individual in crowd images. Recently, several methods have been proposed for semi-supervised crowd counting to reduce the labeling efforts. Given a limited labeling budget, they typically select a few crowd images and densely label all individuals in each of them. Despite the promising results, we argue the None-or-All labeling strategy is suboptimal as the densely labeled individuals in each crowd image usually appear similar while the massive unlabeled crowd images may contain entirely diverse individuals. To this end, we propose to break the labeling chain of previous methods and make the first attempt to reduce spatial labeling redundancy for semi-supervised crowd counting. First, instead of annotating all the regions in each crowd image, we propose to annotate the representative ones only. We analyze the region representativeness from both vertical and horizontal directions of initially estimated density maps, and formulate them as cluster centers of Gaussian Mixture Models. Additionally, to leverage the rich unlabeled regions, we exploit the similarities among individuals in each crowd image to directly supervise the unlabeled regions via feature propagation instead of the error-prone label propagation employed in the previous methods. In this way, we can transfer the original spatial labeling redundancy caused by individual similarities to effective supervision signals on the unlabeled regions. Extensive experiments on the widely-used benchmarks demonstrate that our method can outperform previous best approaches by a large margin.

Automated summary

Labeling is challenging for crowd counting, and recent methods have proposed semi-supervised approaches to reduce labeling efforts. However, the None-or-All labeling strategy is suboptimal as it does not consider the diversity of individuals in unlabeled crowd images. In this study, we propose breaking the labeling chain and reducing spatial labeling redundancy to improve semi-supervised crowd counting. We annotate representative regions, analyze region representativeness, and directly supervise unlabeled regions using similarity among individuals. Our experiments show significant performance improvement compared to previous methods.