{"id":326,"date":"2019-07-01T14:50:57","date_gmt":"2019-07-01T12:50:57","guid":{"rendered":"https:\/\/project.inria.fr\/crowdscience\/?p=326"},"modified":"2020-03-04T10:41:27","modified_gmt":"2020-03-04T09:41:27","slug":"data-driven-crowd-simulation-with-gans-casa-2019","status":"publish","type":"post","link":"https:\/\/project.inria.fr\/crowdscience\/data-driven-crowd-simulation-with-gans-casa-2019\/","title":{"rendered":"Publication (CASA 2019): Data-Driven Crowd Simulation with GANs"},"content":{"rendered":"<p>(This paper has been published in the 2019<em> International Conference on Computer Animation and Social Agents<\/em>.)<\/p>\n<h4>Abstract:<\/h4>\n<p>This paper presents a novel <strong>data-driven crowd simulation method<\/strong> that can mimic the observed traffic of pedestrians in a given environment. Given a set of observed trajectories, we use a recent form of neural networks,<strong> Generative Adversarial Networks (GANs)<\/strong>, to learn the properties of this set and generate <strong>new trajectories<\/strong> with similar properties. We define a way for simulated pedestrians (agents) to follow such a trajectory while handling local collision avoidance. As such, the system can generate a crowd that behaves similarly to observations, while still enabling real-time interactions between agents.<\/p>\n<p>Via experiments with several real-world data sets, we show that our simulated trajectories preserve the statistical properties of their input. Our method can simulate crowds in <strong>real time<\/strong> that <strong>resemble existing crowds<\/strong>, while also allowing the insertion of extra agents, the combination with other simulation methods, and user interaction.<\/p>\n<h4><a href=\"https:\/\/project.inria.fr\/crowdscience\/files\/2019\/12\/casa2019.png\"><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-medium wp-image-328\" src=\"https:\/\/project.inria.fr\/crowdscience\/files\/2019\/12\/casa2019-300x187.png\" alt=\"\" width=\"300\" height=\"187\" srcset=\"https:\/\/project.inria.fr\/crowdscience\/files\/2019\/12\/casa2019-300x187.png 300w, https:\/\/project.inria.fr\/crowdscience\/files\/2019\/12\/casa2019-768x478.png 768w, https:\/\/project.inria.fr\/crowdscience\/files\/2019\/12\/casa2019-1024x637.png 1024w, https:\/\/project.inria.fr\/crowdscience\/files\/2019\/12\/casa2019-150x93.png 150w, https:\/\/project.inria.fr\/crowdscience\/files\/2019\/12\/casa2019.png 1121w\" sizes=\"auto, (max-width: 300px) 100vw, 300px\" \/><\/a><\/h4>\n<h4>Details:<\/h4>\n<ul>\n<li>Title: <strong>Data-Driven Crowd Simulation with Generative Adversarial Networks<\/strong><\/li>\n<li>Authors: Javad Amirian, Wouter Van Toll, Jean-Bernard Hayet, Julien Pettr\u00e9<\/li>\n<li>Download the full article: <a href=\"https:\/\/hal.inria.fr\/hal-02134282\/\">https:\/\/hal.inria.fr\/hal-02134282\/<\/a><\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n","protected":false},"excerpt":{"rendered":"<p>(This paper has been published in the 2019 International Conference on Computer Animation and Social Agents.) Abstract: This paper presents a novel data-driven crowd simulation method that can mimic the observed traffic of pedestrians in a given environment. Given a set of observed trajectories, we use a recent form of\u2026<\/p>\n<p> <a class=\"continue-reading-link\" href=\"https:\/\/project.inria.fr\/crowdscience\/data-driven-crowd-simulation-with-gans-casa-2019\/\"><span>Continue reading<\/span><i class=\"crycon-right-dir\"><\/i><\/a> <\/p>\n","protected":false},"author":1605,"featured_media":328,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[14,5],"tags":[18,20],"class_list":["post-326","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news","category-publications","tag-theme-crowd-simulation","tag-theme-machine-learning"],"_links":{"self":[{"href":"https:\/\/project.inria.fr\/crowdscience\/wp-json\/wp\/v2\/posts\/326","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/project.inria.fr\/crowdscience\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/project.inria.fr\/crowdscience\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/project.inria.fr\/crowdscience\/wp-json\/wp\/v2\/users\/1605"}],"replies":[{"embeddable":true,"href":"https:\/\/project.inria.fr\/crowdscience\/wp-json\/wp\/v2\/comments?post=326"}],"version-history":[{"count":8,"href":"https:\/\/project.inria.fr\/crowdscience\/wp-json\/wp\/v2\/posts\/326\/revisions"}],"predecessor-version":[{"id":584,"href":"https:\/\/project.inria.fr\/crowdscience\/wp-json\/wp\/v2\/posts\/326\/revisions\/584"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/project.inria.fr\/crowdscience\/wp-json\/wp\/v2\/media\/328"}],"wp:attachment":[{"href":"https:\/\/project.inria.fr\/crowdscience\/wp-json\/wp\/v2\/media?parent=326"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/project.inria.fr\/crowdscience\/wp-json\/wp\/v2\/categories?post=326"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/project.inria.fr\/crowdscience\/wp-json\/wp\/v2\/tags?post=326"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}