@article{Luo_Yang_Li_Nie_Jiao_Zhou_Cheng_2020, title={Hybrid Graph Neural Networks for Crowd Counting}, volume={34}, url={https://ojs.aaai.org/index.php/AAAI/article/view/6839}, DOI={10.1609/aaai.v34i07.6839}, abstractNote={<p>Crowd counting is an important yet challenging task due to the large scale and density variation. Recent investigations have shown that distilling rich relations among multi-scale features and exploiting useful information from the auxiliary task, i.e., localization, are vital for this task. Nevertheless, how to comprehensively leverage these relations within a unified network architecture is still a challenging problem. In this paper, we present a novel network structure called Hybrid Graph Neural Network (H<span style="font-variant: small-caps;">y</span>G<span style="font-variant: small-caps;">nn</span>) which targets to relieve the problem by interweaving the multi-scale features for crowd density as well as its auxiliary task (localization) together and performing joint reasoning over a graph. Specifically, H<span style="font-variant: small-caps;">y</span>G<span style="font-variant: small-caps;">nn</span> integrates a hybrid graph to jointly represent the task-specific feature maps of different scales as nodes, and two types of relations as edges: <strong>(i)</strong> multi-scale relations capturing the feature dependencies across scales and <strong>(ii)</strong> mutual beneficial relations building bridges for the cooperation between counting and localization. Thus, through message passing, H<span style="font-variant: small-caps;">y</span>G<span style="font-variant: small-caps;">nn</span> can capture and distill richer relations between nodes to obtain more powerful representations, providing robust and accurate results. Our H<span style="font-variant: small-caps;">y</span>G<span style="font-variant: small-caps;">nn</span> performs significantly well on four challenging datasets: ShanghaiTech Part A, ShanghaiTech Part B, UCF_CC_50 and UCF_QNRF, outperforming the state-of-the-art algorithms by a large margin.</p>}, number={07}, journal={Proceedings of the AAAI Conference on Artificial Intelligence}, author={Luo, Ao and Yang, Fan and Li, Xin and Nie, Dong and Jiao, Zhicheng and Zhou, Shangchen and Cheng, Hong}, year={2020}, month={Apr.}, pages={11693-11700} }