{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/stacked-pooling-improving-crowd-counting-by","title":"Stacked Pooling: Improving Crowd Counting by Boosting Scale Invariance","arxiv_id":"1808.07456","date":"2018-08-22","proceeding":null,"authors":["Siyu Huang","Xi Li","Zhi-Qi Cheng","Zhongfei Zhang","Alexander Hauptmann"],"abstract":"In this work, we explore the cross-scale similarity in crowd counting\nscenario, in which the regions of different scales often exhibit high visual\nsimilarity. This feature is universal both within an image and across different\nimages, indicating the importance of scale invariance of a crowd counting\nmodel. Motivated by this, in this paper we propose simple but effective\nvariants of pooling module, i.e., multi-kernel pooling and stacked pooling, to\nboost the scale invariance of convolutional neural networks (CNNs), benefiting\nmuch the crowd density estimation and counting. Specifically, the multi-kernel\npooling comprises of pooling kernels with multiple receptive fields to capture\nthe responses at multi-scale local ranges. The stacked pooling is an equivalent\nform of multi-kernel pooling, while, it reduces considerable computing cost.\nOur proposed pooling modules do not introduce extra parameters into model and\ncan easily take place of the vanilla pooling layer in implementation. In\nempirical study on two benchmark crowd counting datasets, the stacked pooling\nbeats the vanilla pooling layer in most cases.","url_abs":"http://arxiv.org/abs/1808.07456v1","url_pdf":"http://arxiv.org/pdf/1808.07456v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"stacked-pooling-improving-crowd-counting-by","repo_url":"https://github.com/zhiqic/crowdcount-stackpool","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"crowd-counting","task_name":"Crowd Counting"},{"task_slug":"density-estimation","task_name":"Density Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}