[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-160254-en":3,"doc-seo-160254-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"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":14,"update_tm":28,"read_time":29},160254,5909887254083,"\tWilliam","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","A Statistical Assessment of Subject Factors in the PCA Recognition of Human Subjects - read online free","The material presents a statistical assessment of which subject factors most influence PCA-based recognition of human subjects. It discusses selecting a simple PCA algorithm, using a distance metric such as Yambor Angle, defining and collecting covariates, and formulating which quantities to measure. It outlines core image preprocessing steps—conversion, geometric normalization, masking, histogram equalization, and pixel normalization—and proposes evaluation strategies using partitioned recognition-rate data or intrapersonal image-pair distances with appropriate statistical models.","A Statistical Assessment of Subject Factors in the PCA Recognition of Human Subjects  \nGeof Givens*, J. Ross Beveridge, Bruce A. Draper & David Bolme  \nComputer Science, Colorado State University  \n*Statistics, Colorado State University  \nCVPR Workshop: Statistical Analysis in Computer Vision  \nJune 22, 2003  \n\n|  |  |\n| --- | --- |\n| | Some Human Subjects Are Harder to Recognize than Others. Why? |\n\n|  |  |\n| --- | --- |\n| | Better Formulate the Question |\n\n• Choose an algorithm  \n– Something standard and simple.  \n– PCA using our publicly released version.  \n• Choose a distance measure:  \n– Yambor Angle is good and similar to Moon-Phillips FERET.  \n– Much better than standards such as L2 .  \n• Define & Collect Covariates:  \n– Few covariates were collected initially with FERET images.  \n– One person at CSU, David Bolme, scored all images.  \n• Big questions.  \n– What to measure?  \n– Covariates in isolation or together?  \n\n|  |  |\n| --- | --- |\n| | Some Basics: Image Preprocessing |\n\n• Integer to float conversion  \n– Converts 256 gray levels to singlefloats  \n• Geometric Normalization  \n– Aligns human chosen eye coordinates  \n• Masking  \n– Crop with elliptical mask leaving only face visible.  \n• Histogram Equalization  \n– Histogram equalizes unmasked pixels: 256 levels.  \n• Pixel normalization  \n– Shift and scale pixel values so mean pixel value is zero and standard deviation over all pixels is one.  \nRefinement of NIST preprocessing used in FERET.  \nCVPR Workshop: Statistical Analysis in Computer Vision Page 4  \n\n|  |  |\n| --- | --- |\n| | More Basics: Standard PCA Algorithm . |\n\nTraining  \nTraining images  \n… …  \nEigenspace  \nTesting  \n… …  \nDistance Matrix  \nPCA space projection  \nRemove in/out training issue by training on all images.  \n\n|  |  |\n| --- | --- |\n| | What Covariates? |\n\nRace  \nGender  \nAge  \nEyes  \nGlasses  \nBangs  \nFacial Hair  \nMouth  \nSmiling?  \n\n|  |  |\n| --- | --- |\n| | The Final Set of Subject Covariates |\n\n\n| FERET Subject/Image Covariates |  |\n| --- | --- |\n| Fixed Per Subject |  |\n| Age | Young Old |\n| Gender | Male Female |\n| Race | White Black Asian Other |\n| Skin | Clear Other |\n| Fixed Per Image |  |\n| Bangs | No Yes |\n| Expression | Neutral Other |\n| Eyes | Open Other |\n| Facial Hair | No Yes |\n| Makeup | No Yes |\n| Mouth | Closed Other |\n\n\n|  |  |\n| --- | --- |\n| | Collecting the Covariates |\n\nCVPR Workshop: Statistical Analysis in Computer Vision Page 8  \n\n|  |  |\n| --- | --- |\n| | What to Measure? Recognition Rate on Partitioned Data |\n\n• Measure Recognition Rate for Partitioned Images.  \n– Partition images by covariate : e.g. male versus female.  \n– Compare recognition rate on different sets.  \n• Good  \n– Answers a very specific version of the question.  \n– Recognition rate is a standard performance measure.  \n• Bad  \n– Fails to adjust, i.e. control for, other covariates.  \n– Recognition rate for a probe set, not a single subject.  \n– Hidden dependence on gallery image set.  \n– Statistical significance is hard to interpret.  \n\n|  |  |\n| --- | --- |\n| | What to measure? Intrapersonal Image Pair Distance |\n\n• Measure Distance (Similarity) Between Images  \n– For two images of one subject, closer is better.  \n• Good  \n– Measure is independent of other subjects, probe sets, etc.  \n– Measure is continuous, analysis of variance appropriate.  \n– Linear model accounts for all covariates at one time.  \n– Standard tests of statistical significance apply.  \n• Bad  \n– Distance to recognition rate connection indirect.  \n– Supplemental analysis required to establish linkage.  \n• Made through intermediate measure: recognition rank.","cbCaiokKYVishRrq","https://ap.wps.com/l/cbCaiokKYVishRrq","pdf",2250934,1,26,"English","en",105,"# Algorithm and Distance Choice\n## Covariate Definition and Collection\n# Image Preprocessing Basics\n## Standard PCA Algorithm Overview\n# Covariates and Their Final Set\n## Collecting the Covariates\n# Measurement Strategies for Recognition Quality\n## Recognition Rate on Partitioned Data\n## Intrapersonal Image Pair Distance","[{\"question\":\"Why do some human subjects get recognized less reliably than others?\",\"answer\":\"The document frames recognition difficulty as a function of measurable subject factors (covariates) and asks how to statistically assess their effect on PCA recognition performance.\"},{\"question\":\"What covariates are considered in the final subject/image set?\",\"answer\":\"It lists covariates such as Age, Gender, Race, Skin, Bangs, Expression, Eyes, Facial Hair, Makeup, and Mouth, with guidance on fixed-per-subject versus fixed-per-image attributes.\"},{\"question\":\"What evaluation quantity is recommended: partitioned recognition rate or intrapersonal image-pair distance?\",\"answer\":\"Recognition rate on partitioned data answers a specific question but requires controlling for other covariates and has interpretability issues. Intrapersonal image-pair distance directly measures similarity within the same subject and supports standard statistical modeling, though it may require supplemental analysis to connect to recognition-rate behavior.\"}]","A Statistical Assessment of Subject Factors in the PCA Recognition of Human Subjects - read online free | PDF",1788052815,66,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"a-statistical-assessment-of-subject-factors-in-the-pca-recognition-of-human-subjects-read-online-free","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/a-statistical-assessment-of-subject-factors-in-the-pca-recognition-of-human-subjects-read-online-free/160254/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-30",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why do some human subjects get recognized less reliably than others?","Question",{"text":75,"@type":76},"The document frames recognition difficulty as a function of measurable subject factors (covariates) and asks how to statistically assess their effect on PCA recognition performance.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What covariates are considered in the final subject/image set?",{"text":80,"@type":76},"It lists covariates such as Age, Gender, Race, Skin, Bangs, Expression, Eyes, Facial Hair, Makeup, and Mouth, with guidance on fixed-per-subject versus fixed-per-image attributes.",{"name":82,"@type":73,"acceptedAnswer":83},"What evaluation quantity is recommended: partitioned recognition rate or intrapersonal image-pair distance?",{"text":84,"@type":76},"Recognition rate on partitioned data answers a specific question but requires controlling for other covariates and has interpretability issues. Intrapersonal image-pair distance directly measures similarity within the same subject and supports standard statistical modeling, though it may require supplemental analysis to connect to recognition-rate behavior.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]