Action recognition algorithm based on skeleton graph with multiple features and improved adjacency matrix

Action recognition algorithm based on skeleton graph with multiple features and improved adjacency matrix | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Action recognition algorithm based on skeleton graph with multiple features and improved adjacency matrix Shanqing Zhang, Shuheng Jiao, Yujie Chen, Jiayi Xu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3830147/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Although graph convolutional networks have achieved good performances in skeleton graph-based action recognition, there are still some problems such as: incomplete utilization of skeleton graph features the adjacency matrix without logical adjacency information between nodes. The skeleton graph features and the graph convolutional operators demand further improvement. In this paper, we propose a human action recognition algorithm based on multiple features of the skeleton graph. Furthermore, an improved adjacency matrix make full use of the multiple skeleton graph features. These features include local differential features of the skeleton graph, multi-scale edge features and motion features of the original skeleton graph, in addition to the nodal features and nodal motion features. Extensive results are conducted on four standard datasets (NTU RGB-D 60, NTU RGB-D 120, Kinetics, and Northwestern-UCLA). The experimental results show that the proposed algorithm outperforms the SOTA action recognition algorithms. human action recognition skeleton graph graph convolutional networks multiscale features adjacency matrix topological relations Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3830147","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":264997886,"identity":"35aeb82b-03bd-4433-b62b-66965fd4c7ac","order_by":0,"name":"Shanqing Zhang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAzUlEQVRIiWNgGAWjYHACNhAhZ8CQAGMTqcWYdC2JG4jWIt+e/uzBzx216dvZcwwYPpQdZuCf3YBfi8GZB+mGvWeO5+7seWPAOOPcYQaJOwcIaJFIOCbB23Ysd8ONHANm3rbDIBECDpuR2Cb5t+1YugFIy19itDDcSGaT5m2rSQBrYSRGi8GZZ2zSsm0HDDeceVZwsOdcOo/EDUIOA4aY5Nu2OnmD48kbH/wos5bjn0HIYeDoYDgMZh4AYh5C6mFa6ohQOApGwSgYBSMWAAD830YWWayklgAAAABJRU5ErkJggg==","orcid":"","institution":"Hangzhou Dianzi University","correspondingAuthor":true,"prefix":"","firstName":"Shanqing","middleName":"","lastName":"Zhang","suffix":""},{"id":264997887,"identity":"f59926b8-c698-478a-aa31-cbbb864201ca","order_by":1,"name":"Shuheng Jiao","email":"","orcid":"","institution":"Hangzhou Dianzi University","correspondingAuthor":false,"prefix":"","firstName":"Shuheng","middleName":"","lastName":"Jiao","suffix":""},{"id":264997888,"identity":"90ab9521-6d6e-431b-a880-b73ac9ace127","order_by":2,"name":"Yujie Chen","email":"","orcid":"","institution":"Hangzhou Dianzi University","correspondingAuthor":false,"prefix":"","firstName":"Yujie","middleName":"","lastName":"Chen","suffix":""},{"id":264997889,"identity":"d68bebb9-c080-4e28-ad57-9f14995a6836","order_by":3,"name":"Jiayi Xu","email":"","orcid":"","institution":"Hangzhou Dianzi University","correspondingAuthor":false,"prefix":"","firstName":"Jiayi","middleName":"","lastName":"Xu","suffix":""}],"badges":[],"createdAt":"2024-01-02 18:44:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3830147/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3830147/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":50766337,"identity":"efc4a12a-c13d-4cda-9f9f-c5859786a47c","added_by":"auto","created_at":"2024-02-07 02:46:00","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4767237,"visible":true,"origin":"","legend":"","description":"","filename":"paper.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3830147/v1_covered_12865ba3-5064-4eba-8f93-81021adc2da9.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Action recognition algorithm based on skeleton graph with multiple features and improved adjacency matrix","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"human action recognition, skeleton graph, graph convolutional networks, multiscale features, adjacency matrix, topological relations","lastPublishedDoi":"10.21203/rs.3.rs-3830147/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3830147/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Although graph convolutional networks have achieved good performances in skeleton graph-based action recognition, there are still some problems such as: incomplete utilization of skeleton graph features the adjacency matrix without logical adjacency information between nodes. The skeleton graph features and the graph convolutional operators demand further improvement. In this paper, we propose a human action recognition algorithm based on multiple features of the skeleton graph. Furthermore, an improved adjacency matrix make full use of the multiple skeleton graph features. These features include local differential features of the skeleton graph, multi-scale edge features and motion features of the original skeleton graph, in addition to the nodal features and nodal motion features. Extensive results are conducted on four standard datasets (NTU RGB-D 60, NTU RGB-D 120, Kinetics, and Northwestern-UCLA). The experimental results show that the proposed algorithm outperforms the SOTA action recognition algorithms.","manuscriptTitle":"Action recognition algorithm based on skeleton graph with multiple features and improved adjacency matrix","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-01-05 02:18:53","doi":"10.21203/rs.3.rs-3830147/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"3447927d-07c1-41bb-ad0f-d0867c1bc55e","owner":[],"postedDate":"January 5th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-02-07T02:37:35+00:00","versionOfRecord":[],"versionCreatedAt":"2024-01-05 02:18:53","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3830147","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3830147","identity":"rs-3830147","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}