{"id":49315,"date":"2021-09-13T12:24:53","date_gmt":"2021-09-13T09:24:53","guid":{"rendered":"https:\/\/forklog.com\/en\/?p=49315"},"modified":"2025-09-02T12:06:02","modified_gmt":"2025-09-02T09:06:02","slug":"scientists-explain-how-to-distinguish-a-deepfake-from-a-real-photo","status":"publish","type":"post","link":"https:\/\/u1f987.com\/en\/scientists-explain-how-to-distinguish-a-deepfake-from-a-real-photo\/","title":{"rendered":"Scientists explain how to distinguish a deepfake from a real photo"},"content":{"rendered":"<p>A group of American scientists<a href=\"https:\/\/arxiv.org\/pdf\/2109.00162.pdf\" target=\"_blank\" rel=\"noreferrer noopener nofollow\"> has discovered<\/a> a feature that allows distinguishing deepfakes from photos of real people by examining the eyes.<\/p>\n<p>Researchers found that most generative adversarial networks (GANs) used to create fakes do not draw round pupils. Such artifacts are widespread among high-quality deepfakes, they say.<\/p>\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p>&#8220;The incorrect shape of the pupil is a good sign that a person can identify a face generated by a GAN. Even if there are no boundary marks around the pupils, we will easily notice their irregular shapes, which can vary significantly on the same deepfake image,&#8221; the paper states.<\/p>\n<\/blockquote>\n<figure class=\"wp-block-image\"><img decoding=\"async\" src=\"https:\/\/lh4.googleusercontent.com\/uqw9RjH5Bue3SveJxqSVe2z5akK146PcDZcpl-TFlsXjJb9GYQdLmi4Fs0g9mLHdOEO1eQ_MUpeBWgv7pEIEhjR_mqnE0MqNPN8GoqWKPUYJsnqgY9Z02qF_lFEWlpm7UT_KfYmU=s0\" alt=\"Scientists explain how to tell a deepfake from a real photo\"\/><figcaption>Pupil shape in a real photo and in a deepfake. Data: study.<\/figcaption><\/figure>\n<p>Researchers note that in practice people can enlarge the image and check the pupil shape to detect a forgery.<\/p>\n<p>Back in June, Facebook developed<a href=\"https:\/\/u1f987.com\/en\/news\/facebook-develops-algorithm-to-detect-deepfakes-and-trace-their-source\"> an algorithm for detecting deepfakes<\/a>, which tracks the source of the fabrication.<\/p>\n<p>In April, experts stated that<a href=\"https:\/\/u1f987.com\/en\/news\/experts-the-number-of-deepfakes-online-doubles-roughly-every-six-months\"> the number of online forgeries doubles<\/a> every six months.<\/p>\n<p>In March, scientists developed a tool that<a href=\"https:\/\/u1f987.com\/en\/news\/scientists-devise-tool-to-expose-deepfakes\"> based on reflections of light in the cornea can recognise a deepfake<\/a>.<\/p>\n<p>Follow ForkLog news on Telegram:<a href=\"https:\/\/t.me\/forklogAI\" target=\"_blank\" rel=\"noreferrer noopener nofollow\"> ForkLog AI<\/a> \u2014 all the news from the world of AI!<\/p>\n","protected":false},"excerpt":{"rendered":"<p>A group of American scientists has discovered a feature that can distinguish deepfakes from real photos of people with high probability by examining the eyes.<\/p>\n","protected":false},"author":1,"featured_media":26216,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"select":"1","news_style_id":"1","cryptorium_level":"","_short_excerpt_text":"","creation_source":"","_metatest_mainpost_news_update":false,"footnotes":""},"categories":[3],"tags":[438,1250,167],"class_list":["post-49315","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news-and-analysis","tag-artificial-intelligence","tag-deepfakes","tag-research"],"aioseo_notices":[],"amp_enabled":true,"views":"23","promo_type":"1","layout_type":"1","short_excerpt":"","is_update":"","_links":{"self":[{"href":"https:\/\/u1f987.com\/en\/wp-json\/wp\/v2\/posts\/49315","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/u1f987.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/u1f987.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/u1f987.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/u1f987.com\/en\/wp-json\/wp\/v2\/comments?post=49315"}],"version-history":[{"count":1,"href":"https:\/\/u1f987.com\/en\/wp-json\/wp\/v2\/posts\/49315\/revisions"}],"predecessor-version":[{"id":49316,"href":"https:\/\/u1f987.com\/en\/wp-json\/wp\/v2\/posts\/49315\/revisions\/49316"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/u1f987.com\/en\/wp-json\/wp\/v2\/media\/26216"}],"wp:attachment":[{"href":"https:\/\/u1f987.com\/en\/wp-json\/wp\/v2\/media?parent=49315"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/u1f987.com\/en\/wp-json\/wp\/v2\/categories?post=49315"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/u1f987.com\/en\/wp-json\/wp\/v2\/tags?post=49315"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}