[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-188187-en":3,"doc-seo-188187-105":30,"detail-sidebar-cat-1-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":11,"category_id":12,"category_name":13,"doc_title":14,"doc_description":15,"doc_content":16,"file_id":17,"file_url":18,"file_type":19,"file_size":20,"view_count":11,"is_deleted":4,"is_public":11,"is_downloadable":11,"audit_status":11,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":15,"update_tm":28,"read_time":29},188187,1099514067415,"Rowan","https://ap-avatar.wpscdn.com/avatar/100002539d78ffe74a7?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779092875211072502",1,158,"General","Deformable Templates for Face Recognition","We describe an approach for extracting facial features from images and determining the spatial organization between these features using deformable templates. The method represents the target as a parameterized geometric model and uses a matching measure to quantify how well the model fits the image. Template parameters capture allowable deformations, and probabilistic modeling links variations to recognition. The approach aims to handle viewpoint, expression, lighting, and partial visibility challenges.","# Deformable Templates for Face Recognition\n\nAlan L.Yuille  \nDivision of Applied ScienceHarvard University  \nAbstract  \n■We describe an approach for extracting facial features fromimages and for determining the spatial organization betweenthese features using the concept of a deformable template.Thisis a parameterized geometric model of the object to be rec-ognized together with a measure of how well it fits the image  \n## INTRODUCTION\n\nThe sophistication of the human visual system is oftentaken for granted until we try to design artificial systemswith similar capabilities.In particular humans have anamazing ability to recognize faces seen from differentviewpoints,with various expressions and under a varietyof lighting conditions.It is hoped that current attemptsto build computer vision face recognition systems willshed light on how humans perform this task and thedifficulties they overcome.  \nIn this article we describe one promising approachtoward building a face recognition system.Some alter-native approaches are reviewed in Turk and Pentland(1989).  \nA standard way of describing a face consists of repre-senting the features and the spatial relations betweenthem.Indeed two of the earliest face recognition systems(Goldstein,Harmon,&Lesk,1972;Kanade,1977)usedfeatures and spatial relations,respectively.Such repre-sentations are independent of lighting conditions,maybe able to characterize different spatial expressions,andhave some limited viewpoint invariance.  \nThese representations,however,are extremely difficultto compute reliably from a photograph of a face.It isvery hard to extract features,such as the eyes,by usingcurrent computer vision techniques.Deformable tem-plates,however,offer a promising way of using a prioriknowledge about features,and the spatial relationshipsbetween them,in the detection stage.Once the featurehas been extracted the parameters of the deformabletemplate can be used for description and recognition.  \nDesigning a deformable template to detect a feature,or an object,falls into two parts:(1)providing a geo-metric model for the template,and(2)specifying amodel,the imaging model,for how a template of specific  \ndata.Variations in the parameters correspond to allowable de-formations of the object and can be specified by a probabilis-tic model.After the extraction stage the parameters of thedeformable template can be used for object description andrecognition.  \ngeometry will appear in the image and a correspondingmeasure of fitness,or matching criterion,to determinehow the template interacts with the image.  \nFor many objects it is straightforward to specify a plau-sible geometric model.Intuition is often a useful guideand one can draw on the experience of artists(for ex-ample,Bridgman,1973).Once the form of the model isspecified it can be evaluated and the probability distri-bution of the parameters determined by statistical tests,given enough instances of the object  \nSpecifying the imaging model and the matching cri-terion is often considerably harder.The way the objectreflects light depends both on the reflectance functionof the object,which in principle can be modeled,andon the scene illumination,which is usually unknown.The matching criterion,however,should aim to be rel-atively independent of the lighting conditions.An evenmore serious problem arises when part of the object isinvisible,perhaps due to occlusion by another object orby lying in deep shadow.Most matching criteria willbreak down in this situation but recent work,describedin the fourth section,shows promise of dealing with suchproblems.  \nThe plan of this chapter is as follows.The secondsection gives a simple introduction to deformable tem-plates by describing the pioneering work of Fischler andElschlager(1973)on extracting global descriptions offaces.The third section gives a more detailed descriptionof using deformable templates to extract facial features(Yuille,Cohen,Hallinan,1989).The fourth section showsa way to put deformable te","cbCaigJiCLmZTnf4","https://ap.wps.com/l/cbCaigJiCLmZTnf4","pdf",4110605,12,"English","en",105,"# INTRODUCTION\n# GLOBAL TEMPLATES","[{\"question\":\"What are deformable templates used for in face recognition?\",\"answer\":\"They are used to extract facial features and determine the spatial relationships among those features. The template is a parameterized geometric model fitted to the image using a fitness or matching measure.\"},{\"question\":\"How does the global template approach model a face?\",\"answer\":\"It models the face as a set of basic, non-deforming features connected by springs. During matching, the structure deforms like a rubber sheet until local fits are good and spring forces are balanced.\"},{\"question\":\"Why is specifying the imaging model and matching criterion challenging?\",\"answer\":\"Because reflectance depends on the object and illumination, and illumination is usually unknown. Matching criteria also must cope with partial invisibility from occlusion or deep shadow, where many criteria break down.\"}]","Deformable Templates for Face Recognition | PDF",1788387750,4,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":14,"keywords":34,"description":15,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"deformable-templates-for-face-recognition","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":11},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/template/","Template",2,{"item":49,"name":13,"@type":43,"position":50},"https://docshare.wps.com/template/general/",3,{"item":52,"name":14,"@type":43,"position":29},"https://docshare.wps.com/template/deformable-templates-for-face-recognition/188187/",{"url":52,"name":14,"@type":54,"author":55,"headline":14,"publisher":57,"fileFormat":60,"inLanguage":23,"description":15,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-09-06","2026-09-02",true,{"@type":65,"interactionType":66,"userInteractionCount":11},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What are deformable templates used for in face recognition?","Question",{"text":75,"@type":76},"They are used to extract facial features and determine the spatial relationships among those features. The template is a parameterized geometric model fitted to the image using a fitness or matching measure.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the global template approach model a face?",{"text":80,"@type":76},"It models the face as a set of basic, non-deforming features connected by springs. During matching, the structure deforms like a rubber sheet until local fits are good and spring forces are balanced.",{"name":82,"@type":73,"acceptedAnswer":83},"Why is specifying the imaging model and matching criterion challenging?",{"text":84,"@type":76},"Because reflectance depends on the object and illumination, and illumination is usually unknown. Matching criteria also must cope with partial invisibility from occlusion or deep shadow, where many criteria break down.","https://schema.org",{"og:url":52,"og:type":87,"og:title":14,"og:site_name":58,"og:description":15},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,98,102,107,112,117,122,127,132],{"id":94,"doc_module":11,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},11,"Presentations",90,"presentations",{"id":21,"doc_module":11,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Resumes",80,"resumes",{"id":103,"doc_module":11,"doc_module_name":46,"category_name":104,"show_sort_weight":105,"slug":106},14,"Invoices",70,"invoices",{"id":108,"doc_module":11,"doc_module_name":46,"category_name":109,"show_sort_weight":110,"slug":111},15,"Posters",60,"posters",{"id":113,"doc_module":11,"doc_module_name":46,"category_name":114,"show_sort_weight":115,"slug":116},16,"Social Media",50,"social-media",{"id":118,"doc_module":11,"doc_module_name":46,"category_name":119,"show_sort_weight":120,"slug":121},17,"Forms",40,"forms",{"id":123,"doc_module":11,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},18,"Letters",30,"letters",{"id":128,"doc_module":11,"doc_module_name":46,"category_name":129,"show_sort_weight":130,"slug":131},21,"Paper Templates",5,"papers-templates",{"id":12,"doc_module":11,"doc_module_name":46,"category_name":13,"show_sort_weight":4,"slug":133},"general-158"]