CascadeNet: A Two-Stage Hybrid Learning Framework for Explainable Deepfake Forensics

CascadeNet: A Two-Stage Hybrid Learning Framework for Explainable Deepfake Forensics | 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 CascadeNet: A Two-Stage Hybrid Learning Framework for Explainable Deepfake Forensics Jatin Yadav, Divyanshu Sharma, Pritee Khanna, Neha Gour This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8823156/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 This paper presents a novel two-stage deepfake detection framework that helpsin finding the exact locations of the region of the image where a deepfake hashappened. In the first stage, a binary classification approach is used to find outwhether an image or video frame has undergone any manipulation or not. Thisclassifier achieves a high accuracy of 99.9% in classifying whether the imageis deepfake or not while maintaining computational efficiency with the help oftransfer learning and data augmentation methods. After the image is successfullyclassified as a deepfake, the second stage uses a U-Net segmentation model with aResNet encoder to find out the exact regions within each frame where a deepfakehas happened, providing pixel-level manipulation boundaries. Extensive experi-ments on the challenging FF++ dataset demonstrate 99.9% accuracy in detectionand 99.8% Receiver Operating Characteristic - Area Under Curve (ROC-AUC)Score in localization tasks, with minimal false positives. The proposed architec-ture significantly decreases computational costs and time by specifically choosingonly those frames that are deepfake and using a pre-trained model to helpreduce time. The model’s efficacy is also evaluated using various post-processingtechniques. Deepfake Detection Image Manipulation Deep Learning EfficientNet UNet Semantic Segmentation Facial Forgery Media Forensics Two-Stage Detection FF++ Dataset Facial Manipulation Localization Image Authentication 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-8823156","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":588744629,"identity":"48c8b972-8d04-45e1-970b-9ae88d564977","order_by":0,"name":"Jatin Yadav","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8ElEQVRIie3RMYvCMBTA8VcetEs0a8Qv8UBwEv0qSqDTDU7ipkdBRz+MII4pGVz0/AAuVw6cHOp2goMvBSdpTzfh8h9SKPnRlwbA53vDJApeSaHEYGogBCxe9ytIY+7IsBM1EibmGUJ7R/JYkuEnkyeytfT8S7bZ2uCnzUerCCL7Ddm6XARJXTcF2VbbusG+DggiJhhsywmiaCOQ1QVJZ0zgA2AwKychk4AHmyyTO5GnaiKYgKAYCe9E/fEVhcVZOqjcYFs+S6iOZKpIb7/jP3blq1xsfvLx6KCl1Fl2qSAPaXc15gUA0H1pt8/n8/2LbkWeTtIFG/dgAAAAAElFTkSuQmCC","orcid":"","institution":"Indian Institute of Information Technology Design and Manufacturing Jabalpur","correspondingAuthor":true,"prefix":"","firstName":"Jatin","middleName":"","lastName":"Yadav","suffix":""},{"id":588744630,"identity":"891cd1cf-8fc6-4388-84db-629b6ffe45f8","order_by":1,"name":"Divyanshu Sharma","email":"","orcid":"","institution":"Indian Institute of Information Technology Design and Manufacturing Jabalpur","correspondingAuthor":false,"prefix":"","firstName":"Divyanshu","middleName":"","lastName":"Sharma","suffix":""},{"id":588744631,"identity":"8cad645d-6603-44a6-b8cc-c75a9c60470d","order_by":2,"name":"Pritee Khanna","email":"","orcid":"","institution":"Indian Institute of Information Technology Design and Manufacturing Jabalpur","correspondingAuthor":false,"prefix":"","firstName":"Pritee","middleName":"","lastName":"Khanna","suffix":""},{"id":588744632,"identity":"444e78f4-0cb9-46f9-bfc4-7d5783c2ba54","order_by":3,"name":"Neha Gour","email":"","orcid":"","institution":"Indian Institute of Information Technology Design and Manufacturing Jabalpur","correspondingAuthor":false,"prefix":"","firstName":"Neha","middleName":"","lastName":"Gour","suffix":""}],"badges":[],"createdAt":"2026-02-08 16:53:26","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8823156/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8823156/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106960413,"identity":"6c863088-a7db-46cb-b576-f9573543513b","added_by":"auto","created_at":"2026-04-15 09:20:58","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3926052,"visible":true,"origin":"","legend":"","description":"","filename":"SpringerDeepfakeDetection.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8823156/v1_covered_3f397716-72af-4966-a3ab-ddb02da4356e.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"CascadeNet: A Two-Stage Hybrid Learning Framework for Explainable Deepfake Forensics","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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":"Deepfake Detection, Image Manipulation, Deep Learning, EfficientNet, UNet, Semantic Segmentation, Facial Forgery, Media Forensics, Two-Stage Detection, FF++ Dataset, Facial Manipulation Localization, Image Authentication","lastPublishedDoi":"10.21203/rs.3.rs-8823156/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8823156/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"This paper presents a novel two-stage deepfake detection framework that helpsin finding the exact locations of the region of the image where a deepfake hashappened. In the first stage, a binary classification approach is used to find outwhether an image or video frame has undergone any manipulation or not. Thisclassifier achieves a high accuracy of 99.9% in classifying whether the imageis deepfake or not while maintaining computational efficiency with the help oftransfer learning and data augmentation methods. After the image is successfullyclassified as a deepfake, the second stage uses a U-Net segmentation model with aResNet encoder to find out the exact regions within each frame where a deepfakehas happened, providing pixel-level manipulation boundaries. Extensive experi-ments on the challenging FF++ dataset demonstrate 99.9% accuracy in detectionand 99.8% Receiver Operating Characteristic - Area Under Curve (ROC-AUC)Score in localization tasks, with minimal false positives. The proposed architec-ture significantly decreases computational costs and time by specifically choosingonly those frames that are deepfake and using a pre-trained model to helpreduce time. The model’s efficacy is also evaluated using various post-processingtechniques.","manuscriptTitle":"CascadeNet: A Two-Stage Hybrid Learning Framework for Explainable Deepfake Forensics","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-13 06:35:24","doi":"10.21203/rs.3.rs-8823156/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":"16c0db96-80b0-4c3b-a317-2b67c038af95","owner":[],"postedDate":"April 13th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-13T06:35:24+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-13 06:35:24","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8823156","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8823156","identity":"rs-8823156","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}