{"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/improving-dense-crowd-counting-convolutional","title":"Improving Dense Crowd Counting Convolutional Neural Networks using Inverse k-Nearest Neighbor Maps and Multiscale Upsampling","arxiv_id":"1902.05379","date":"2019-01-31","proceeding":null,"authors":["Greg Olmschenk","Hao Tang","Zhigang Zhu"],"abstract":"Gatherings of thousands to millions of people frequently occur for an\nenormous variety of events, and automated counting of these high-density crowds\nis useful for safety, management, and measuring significance of an event. In\nthis work, we show that the regularly accepted labeling scheme of crowd density\nmaps for training deep neural networks is less effective than our alternative\ninverse k-nearest neighbor (i$k$NN) maps, even when used directly in existing\nstate-of-the-art network structures. We also provide a new network architecture\nMUD-i$k$NN, which uses multi-scale upsampling via transposed convolutions to\ntake full advantage of the provided i$k$NN labeling. This upsampling combined\nwith the i$k$NN maps further improves crowd counting accuracy. Our new network\narchitecture performs favorably in comparison with the state-of-the-art.\nHowever, our labeling and upsampling techniques are generally applicable to\nexisting crowd counting architectures.","url_abs":"http://arxiv.org/abs/1902.05379v3","url_pdf":"http://arxiv.org/pdf/1902.05379v3.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":"improving-dense-crowd-counting-convolutional","repo_url":"https://github.com/golmschenk/sr-gan","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"crowd-counting","task_name":"Crowd Counting"},{"task_slug":"management","task_name":"Management"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1902.05379","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}