{"version":"1.0","provider_name":"Rainbow Dash Network","provider_url":"http:\/\/rainbowdash.net\/","type":"link","title":"adiwan (adiwan)'s status on Tuesday, 23-Apr-19 16:45:11 UTC","author_name":"adiwan (adiwan)","author_url":"http:\/\/rainbowdash.net\/adiwan","url":"http:\/\/rainbowdash.net\/notice\/5454373","html":"@<span class=\"vcard\"><a href=\"http:\/\/rainbowdash.net\/user\/60\" class=\"url\" title=\"Scribus\"><span class=\"fn nickname mention\">scribus<\/span><\/a><\/span> Look up &quot;eigenfaces&quot;. It's an application of &quot;eigenimages&quot; that use principal component analysis to pictures of human faces. In the process it generates an image that is an &quot;average&quot; image when each image of the set contributes the same amount to the final image. That can be used to recognize faces or morph faces.<br \/>The quality depends solely on the image set. The more faces are not aligned the &quot;blurrier&quot; it is and the more nightmarish it becomes.<br \/><a href=\"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/6\/67\/Eigenfaces.png\" title=\"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/6\/67\/Eigenfaces.png\" class=\"attachment\" id=\"attachment-874312\" rel=\"nofollow external\">https:\/\/upload.wikimedia.org\/wikipedia\/commons\/6\/67\/Eigenfaces.png<\/a>"}