MNI-GAIR: Multi-scale Normal Image and Grid Attention-based Image Recognition

preprint OA: closed
Full text JSON View at publisher

Abstract

Abstract Achieving high-precision facial recognition in complex scenarios is one of the key challenges in the field of computer vision. This paper proposes a multi-modal collaborative recognition framework, MNI-GAIR, which combines multi-scale normal image generation, dynamic grid attention mechanisms, and point cloud generalization techniques to address the low efficiency of cross-modal feature alignment and the insufficient monocular real-time performance in existing methods. These innovations significantly enhance recognition performance in complex scenarios such as occlusion and extreme poses. Firstly, a multi-scale normal map generation module based on differentiable rendering is designed, which combines GLCM-LBP features and Cascaded Atrous Pyramid (CAP), improving noise robustness by 23.6% bib1. Secondly, a dynamic grid partitioning attention network (DGPA-Net) is proposed, which optimizes grid structures through gradient-driven approaches and incorporates dual-path attention mechanisms, improving recognition accuracy for extreme side-face (\(>75°\)) scenarios by 14.7% bib2. Lastly, a point cloud generalization framework based on Lie group theory is introduced, enabling cross-modal feature fusion and reduces cross-pose error rates (EER) to 1.23% bib3. Experimental results on multiple standard datasets, including FaceScape and LFW, demonstrate that MNI-GAIR outperforms existing methods in terms of accuracy, robustness, and computational efficiency, providing a systematic solution for 3D facial analysis. The source code is available on GitHub at \href{https://github.com/LLxuLL/MNI-GAIR-Multi-scale-Normal-Image-and-Grid-Attention-based-Image-Recognition}{LLxuLL/MNI-GAIR-Multi-scale-Normal-Image-and-Grid-Attention-based-Image-Recognition: Multi-scale Normal Image and Grid Attention-based Image Recognition}.
Full text 27,220 characters · extracted from preprint-html · click to expand
MNI-GAIR: Multi-scale Normal Image and Grid Attention-based Image Recognition | 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 MNI-GAIR: Multi-scale Normal Image and Grid Attention-based Image Recognition Maoyang Xu, Zhuqing Zheng, Borun He, Yinfeng Chen, Jinye Wang, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6821913/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 Achieving high-precision facial recognition in complex scenarios is one of the key challenges in the field of computer vision. This paper proposes a multi-modal collaborative recognition framework, MNI-GAIR, which combines multi-scale normal image generation, dynamic grid attention mechanisms, and point cloud generalization techniques to address the low efficiency of cross-modal feature alignment and the insufficient monocular real-time performance in existing methods. These innovations significantly enhance recognition performance in complex scenarios such as occlusion and extreme poses. Firstly, a multi-scale normal map generation module based on differentiable rendering is designed, which combines GLCM-LBP features and Cascaded Atrous Pyramid (CAP), improving noise robustness by 23.6% bib1 . Secondly, a dynamic grid partitioning attention network (DGPA-Net) is proposed, which optimizes grid structures through gradient-driven approaches and incorporates dual-path attention mechanisms, improving recognition accuracy for extreme side-face (\(>75°\)) scenarios by 14.7% bib2 . Lastly, a point cloud generalization framework based on Lie group theory is introduced, enabling cross-modal feature fusion and reduces cross-pose error rates (EER) to 1.23% bib3 . Experimental results on multiple standard datasets, including FaceScape and LFW, demonstrate that MNI-GAIR outperforms existing methods in terms of accuracy, robustness, and computational efficiency, providing a systematic solution for 3D facial analysis. The source code is available on GitHub at \href{https://github.com/LLxuLL/MNI-GAIR-Multi-scale-Normal-Image-and-Grid-Attention-based-Image-Recognition}{LLxuLL/MNI-GAIR-Multi-scale-Normal-Image-and-Grid-Attention-based-Image-Recognition: Multi-scale Normal Image and Grid Attention-based Image Recognition}. keyword1 Keyword2 Keyword3 Keyword4 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-6821913","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":542329460,"identity":"b5ea53d3-a31c-4ce5-9dee-62d95436c768","order_by":0,"name":"Maoyang Xu","email":"","orcid":"","institution":"Century College, Beijing University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Maoyang","middleName":"","lastName":"Xu","suffix":""},{"id":542329461,"identity":"8a13a7cc-d446-43b5-8af5-cfe9da744f9d","order_by":1,"name":"Zhuqing Zheng","email":"","orcid":"","institution":"Century College, Beijing University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Zhuqing","middleName":"","lastName":"Zheng","suffix":""},{"id":542329462,"identity":"0ea3c917-76cb-4493-97c2-cb98991c2987","order_by":2,"name":"Borun He","email":"","orcid":"","institution":"Century College, Beijing University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Borun","middleName":"","lastName":"He","suffix":""},{"id":542329463,"identity":"62daea55-8907-4226-9a74-0cf57a97c53d","order_by":3,"name":"Yinfeng Chen","email":"","orcid":"","institution":"Century College, Beijing University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Yinfeng","middleName":"","lastName":"Chen","suffix":""},{"id":542329464,"identity":"3fa6f2ac-8dd2-4605-9831-e97e20d2d0ba","order_by":4,"name":"Jinye Wang","email":"","orcid":"","institution":"Century College, Beijing University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Jinye","middleName":"","lastName":"Wang","suffix":""},{"id":542329465,"identity":"49b4e0ae-3c2d-4322-a8a5-f90b6d033e23","order_by":5,"name":"Chen Han","email":"","orcid":"","institution":"Century College, Beijing University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Chen","middleName":"","lastName":"Han","suffix":""},{"id":542329466,"identity":"afe5b019-bc86-4492-9127-01c253ac931f","order_by":6,"name":"Gengyifan Shang","email":"","orcid":"","institution":"Century College, Beijing University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Gengyifan","middleName":"","lastName":"Shang","suffix":""},{"id":542329467,"identity":"f70188bf-3c19-4fca-889d-29c3a5ba7185","order_by":7,"name":"Lihe Chen","email":"","orcid":"","institution":"Century College, Beijing University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Lihe","middleName":"","lastName":"Chen","suffix":""},{"id":542329468,"identity":"c9483a16-a55f-4390-8fae-4fc94b9d379a","order_by":8,"name":"Wancheng Zhao","email":"","orcid":"","institution":"Century College, Beijing University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Wancheng","middleName":"","lastName":"Zhao","suffix":""},{"id":542329469,"identity":"b0082c8b-1eb8-4fe9-a7a5-b970354d246b","order_by":9,"name":"YuFei Zhou","email":"","orcid":"","institution":"Century College, Beijing University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"YuFei","middleName":"","lastName":"Zhou","suffix":""},{"id":542329470,"identity":"c40f9fe4-0d54-4f7d-a2a8-407111bd9868","order_by":10,"name":"Changjiang Zhang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABFElEQVRIiWNgGAWjYDACCTB5AEQkgEg5BgYeErUYQ7QkEKcFDBIbCGmRn9387OGXmjvy5vwLHh74uaM2vb/97DHpwh8M8uY4tDDOOWZuLHPsmeHOGQ8SDvaeOZ4740xemvSMBAbDnQ3YtTBLJJhJSzYcZtxw40DCAd62Y7kbJHjMpHkSGBIMDmDXwiaR/g2kxR6k5eDftmPpBoS08EjkmEl+bDicuOF8Q8Jh3raaBIJaJCRyyqQZjh1O3nCDIeGwbNsBQ6Bfkq1npEkYbsChRX5G+jbJHzWHbTecP5P88W1bnTx/+9mDtwtsbORx2QIOAnB0S+QkAMnDEBFYfOECjD9AJP9xkKl1MC2jYBSMglEwCuAAAK2VZPY4JpVtAAAAAElFTkSuQmCC","orcid":"","institution":"Century College, Beijing University of Posts and Telecommunications","correspondingAuthor":true,"prefix":"","firstName":"Changjiang","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2025-06-04 15:23:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6821913/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6821913/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96321542,"identity":"7593be7d-1de3-4a89-829e-1464a3214fa1","added_by":"auto","created_at":"2025-11-19 19:17:19","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11345,"visible":true,"origin":"","legend":"","description":"","filename":"37944eee1aca4777811437adaefeb6d5.json","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/841a3114c7799c99916120e8.json"},{"id":96366726,"identity":"e4bfec7f-31d3-41b1-80e5-84207b1d7499","added_by":"auto","created_at":"2025-11-20 10:11:49","extension":"xml","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":151854,"visible":true,"origin":"","legend":"","description":"","filename":"37944eee1aca4777811437adaefeb6d51enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/2c7c1787502b82ab5d1775c8.xml"},{"id":96321543,"identity":"7550597f-c864-4abc-9a5f-ae7e9a5bf784","added_by":"auto","created_at":"2025-11-19 19:17:19","extension":"md","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3534,"visible":true,"origin":"","legend":"","description":"","filename":"Coverletter.md","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/5aca73e62aec9991d5bb7fbb.md"},{"id":96321544,"identity":"947c8ce2-d1bd-41b7-8e82-6c665f0a223a","added_by":"auto","created_at":"2025-11-19 19:17:19","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":818368,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/aef5fe5ab692eab75a2e0679.png"},{"id":96321545,"identity":"6ba0fe9a-5c84-4fdc-8bc4-116baad41454","added_by":"auto","created_at":"2025-11-19 19:17:19","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1028645,"visible":true,"origin":"","legend":"","description":"","filename":"Figure10.png","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/549af619376935e78a71f658.png"},{"id":96321547,"identity":"50b57101-f45d-4928-8e3e-27d7aa7ca025","added_by":"auto","created_at":"2025-11-19 19:17:19","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":298133,"visible":true,"origin":"","legend":"","description":"","filename":"Figure11.png","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/c7c0a7b089ef1b731eefa93f.png"},{"id":96366414,"identity":"5e23faa3-52d6-4ffc-87ca-990f5c2a2e77","added_by":"auto","created_at":"2025-11-20 10:11:26","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1966310,"visible":true,"origin":"","legend":"","description":"","filename":"Figure12.png","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/176088e80b878da074e9e326.png"},{"id":96366439,"identity":"15e48385-8127-4ecf-a696-b2fc245a8aa9","added_by":"auto","created_at":"2025-11-20 10:11:27","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3419282,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/013467086c0666d17571eb49.png"},{"id":96321557,"identity":"28dafe1a-ee53-4552-87f7-da102fb1b17f","added_by":"auto","created_at":"2025-11-19 19:17:19","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3304999,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/3fd1b69df2f7ef5742614fc0.png"},{"id":96365562,"identity":"1ef96e01-3b39-4756-9be6-6a67bef5623c","added_by":"auto","created_at":"2025-11-20 10:10:32","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":687804,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/bc5ef73d85d61da10955f07e.png"},{"id":96321550,"identity":"bfc1cd3a-4093-4b78-b1b9-06e962610d23","added_by":"auto","created_at":"2025-11-19 19:17:19","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":665092,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/aa73c64a0c6e6a134d3fa0cf.png"},{"id":96366334,"identity":"7e43637a-3d38-44c7-9ce8-e66725574249","added_by":"auto","created_at":"2025-11-20 10:11:23","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":883446,"visible":true,"origin":"","legend":"","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/26af34f20f74aee80cadc378.png"},{"id":96321569,"identity":"84f25db4-c417-46a3-8c1f-560f17248139","added_by":"auto","created_at":"2025-11-19 19:17:20","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":441928,"visible":true,"origin":"","legend":"","description":"","filename":"Figure7.png","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/77d7829727179006c9febd11.png"},{"id":96365931,"identity":"746cd4e9-f168-4d95-98f4-56fb2c5c90e3","added_by":"auto","created_at":"2025-11-20 10:10:57","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":584241,"visible":true,"origin":"","legend":"","description":"","filename":"Figure8.png","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/f64d49442fb8738634140ccc.png"},{"id":96321554,"identity":"0bacc5bc-be68-4943-8854-45ff9c91870d","added_by":"auto","created_at":"2025-11-19 19:17:19","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1461775,"visible":true,"origin":"","legend":"","description":"","filename":"Figure9.png","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/feee569118090e08ed83469b.png"},{"id":96321572,"identity":"b4d4ecef-3958-4149-8eda-378f9fc30f3c","added_by":"auto","created_at":"2025-11-19 19:17:20","extension":"pdf","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":14706532,"visible":true,"origin":"","legend":"","description":"","filename":"MNIGAIRLaTex.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/fb28e59a141c76556efa614f.pdf"},{"id":96366771,"identity":"0e8c0599-2845-46d4-b30b-782d5ebaf0d4","added_by":"auto","created_at":"2025-11-20 10:11:53","extension":"eps","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2890,"visible":true,"origin":"","legend":"","description":"","filename":"empty.eps","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/b978dff1123851220093f583.eps"},{"id":96321561,"identity":"95f19697-6c02-4533-b28b-4d4962402901","added_by":"auto","created_at":"2025-11-19 19:17:20","extension":"eps","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":91593,"visible":true,"origin":"","legend":"","description":"","filename":"fig.eps","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/5c8bfed89ff215b58df688de.eps"},{"id":96365309,"identity":"16a2abb2-1be6-4342-8103-8737f51ad3b8","added_by":"auto","created_at":"2025-11-20 10:10:15","extension":"bst","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":146013,"visible":true,"origin":"","legend":"","description":"","filename":"snapacite.bst","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/5d7bfaf1c9272c5ff14dd25a.bst"},{"id":96366115,"identity":"067d7115-16d6-4231-a107-71205be75110","added_by":"auto","created_at":"2025-11-20 10:11:12","extension":"bst","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29828,"visible":true,"origin":"","legend":"","description":"","filename":"snaps.bst","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/48e11e1de95feb2b2d732047.bst"},{"id":96321568,"identity":"ff7576a4-f7ac-4a2b-9c89-b6c8d058027e","added_by":"auto","created_at":"2025-11-19 19:17:20","extension":"pdf","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":421391,"visible":true,"origin":"","legend":"","description":"","filename":"snarticle.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/2b096d9d8fc9cc466ee00ca8.pdf"},{"id":96321558,"identity":"d530ced8-487c-4ba2-be38-a7e40faf2cf7","added_by":"auto","created_at":"2025-11-19 19:17:19","extension":"bst","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":35515,"visible":true,"origin":"","legend":"","description":"","filename":"snbasic.bst","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/123e9a769e7b6e147845e2b0.bst"},{"id":96365323,"identity":"1afa54c2-0eb2-4bc9-8fa8-473b26a038f2","added_by":"auto","created_at":"2025-11-20 10:10:15","extension":"bst","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33968,"visible":true,"origin":"","legend":"","description":"","filename":"snchicago.bst","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/26b034c25ca011772e5b0e45.bst"},{"id":96321563,"identity":"b95e7ae5-4016-4f74-b236-6b699c9b4d0e","added_by":"auto","created_at":"2025-11-19 19:17:20","extension":"cls","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":55857,"visible":true,"origin":"","legend":"","description":"","filename":"snjnl.cls","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/3a117b806b278f785b0730dd.cls"},{"id":96367202,"identity":"be1d195d-d3d2-4132-9b24-fbf7f3441eac","added_by":"auto","created_at":"2025-11-20 10:12:16","extension":"bst","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":64023,"visible":true,"origin":"","legend":"","description":"","filename":"snmathphysay.bst","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/c34902fd77e33dd809727b2e.bst"},{"id":96321564,"identity":"0d1803cc-9699-4e8e-87cb-342b7958b878","added_by":"auto","created_at":"2025-11-19 19:17:20","extension":"bst","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":64166,"visible":true,"origin":"","legend":"","description":"","filename":"snmathphysnum.bst","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/eb301a6ad260e39570f84c06.bst"},{"id":96321565,"identity":"fb8c9803-6f4b-4755-b46b-3f37634d16e5","added_by":"auto","created_at":"2025-11-19 19:17:20","extension":"bst","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":37333,"visible":true,"origin":"","legend":"","description":"","filename":"snnature.bst","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/45eff29ad5b62519efbc7a05.bst"},{"id":96321570,"identity":"e69dce86-a713-4b8c-acb4-336e7bf42d3f","added_by":"auto","created_at":"2025-11-19 19:17:20","extension":"bst","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":39951,"visible":true,"origin":"","legend":"","description":"","filename":"snvancouveray.bst","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/c965305ed3429b03af2fb390.bst"},{"id":96321566,"identity":"8eeb2d09-b2d4-4c45-b81d-3795c574d010","added_by":"auto","created_at":"2025-11-19 19:17:20","extension":"bst","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":40758,"visible":true,"origin":"","legend":"","description":"","filename":"snvancouvernum.bst","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/4aec78ecb8670974d123f5f4.bst"},{"id":96367284,"identity":"6c21c403-edf1-47b5-bb10-3cdc4274e46c","added_by":"auto","created_at":"2025-11-20 10:12:29","extension":"pdf","order_by":29,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":418495,"visible":true,"origin":"","legend":"","description":"","filename":"usermanual.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/8fb17e62c55bb268c0fdd476.pdf"},{"id":96321559,"identity":"409cc954-f670-41c6-b9c9-82916b5dca5e","added_by":"auto","created_at":"2025-11-19 19:17:20","extension":"xml","order_by":42,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":161029,"visible":true,"origin":"","legend":"","description":"","filename":"37944eee1aca4777811437adaefeb6d51structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/75b7056e5f16e207f59121c2.xml"},{"id":96321571,"identity":"35b94399-2f18-4bed-a291-12f9da6841f1","added_by":"auto","created_at":"2025-11-19 19:17:20","extension":"html","order_by":43,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":181481,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1/61129a77dd67db63d00da45b.html"},{"id":96598757,"identity":"c69e6aee-8225-4f1a-aeb3-eb34dc9629ec","added_by":"auto","created_at":"2025-11-24 08:09:05","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2526918,"visible":true,"origin":"","legend":"","description":"","filename":"MNIGAIRLaTex.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6821913/v1_covered_a923c28d-90ce-4c6f-9e6b-7e925dcb656d.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"MNI-GAIR: Multi-scale Normal Image and Grid Attention-based Image Recognition","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":"keyword1, Keyword2, Keyword3, Keyword4","lastPublishedDoi":"10.21203/rs.3.rs-6821913/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6821913/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAchieving high-precision facial recognition in complex scenarios is one of the key challenges in the field of computer vision. This paper proposes a multi-modal collaborative recognition framework, MNI-GAIR, which combines multi-scale normal image generation, dynamic grid attention mechanisms, and point cloud generalization techniques to address the low efficiency of cross-modal feature alignment and the insufficient monocular real-time performance in existing methods. These innovations significantly enhance recognition performance in complex scenarios such as occlusion and extreme poses. Firstly, a multi-scale normal map generation module based on differentiable rendering is designed, which combines GLCM-LBP features and Cascaded Atrous Pyramid (CAP), improving noise robustness by 23.6% \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003ebib1\u003c/span\u003e. Secondly, a dynamic grid partitioning attention network (DGPA-Net) is proposed, which optimizes grid structures through gradient-driven approaches and incorporates dual-path attention mechanisms, improving recognition accuracy for extreme side-face (\\(\u0026gt;75\u0026deg;\\)) scenarios by 14.7% \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003ebib2\u003c/span\u003e. Lastly, a point cloud generalization framework based on Lie group theory is introduced, enabling cross-modal feature fusion and reduces cross-pose error rates (EER) to 1.23% \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003ebib3\u003c/span\u003e. Experimental results on multiple standard datasets, including FaceScape and LFW, demonstrate that MNI-GAIR outperforms existing methods in terms of accuracy, robustness, and computational efficiency, providing a systematic solution for 3D facial analysis. The source code is available on GitHub at \\href{https://github.com/LLxuLL/MNI-GAIR-Multi-scale-Normal-Image-and-Grid-Attention-based-Image-Recognition}{LLxuLL/MNI-GAIR-Multi-scale-Normal-Image-and-Grid-Attention-based-Image-Recognition: Multi-scale Normal Image and Grid Attention-based Image Recognition}.\u003c/p\u003e","manuscriptTitle":"MNI-GAIR: Multi-scale Normal Image and Grid Attention-based Image Recognition","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-19 19:17:14","doi":"10.21203/rs.3.rs-6821913/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":"d5caec1d-f86d-43cc-9104-6cf843f4f10e","owner":[],"postedDate":"November 19th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-11-24T08:08:16+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-19 19:17:14","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6821913","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6821913","identity":"rs-6821913","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}

Text is read by the "Ask this paper" AI Q&A widget below. Extraction quality varies by source — PMC NXML preserves structure cleanly, OA-HTML may include some navigation residue, and OA-PDF can have broken hyphenation. The publisher copy (via DOI) is the canonical version.

My notes (saved in your browser only)

Ask this paper AI returns verbatim quotes from the full text · source: preprint-html

Answers must be backed by verbatim quotes from this paper's full text. Hallucinated quotes are dropped automatically; if no verbatim passage answers the question, we say so. How this works

Citation neighborhood (no data yet)

We don't have any in-corpus citations linked to this paper yet. This is a recent paper (2025) — citers typically take a year or two to land, and the OpenAlex reference graph may still be filling in.

Source provenance

europepmc
last seen: 2026-05-20T01:45:00.602351+00:00