{"id":11562,"date":"2019-05-25T11:24:21","date_gmt":"2019-05-25T10:24:21","guid":{"rendered":"http:\/\/newsforher.com\/?p=11562"},"modified":"2019-05-25T11:24:24","modified_gmt":"2019-05-25T10:24:24","slug":"watch-samsungs-new-ai-turn-mona-lisa-into-a-realistic-talking-head","status":"publish","type":"post","link":"https:\/\/newsforher.com\/?p=11562","title":{"rendered":"Watch Samsung&#039;s new AI turn Mona Lisa into a realistic talking head"},"content":{"rendered":"\n<!-- Quick Adsense WordPress Plugin: http:\/\/quickadsense.com\/ -->\n<div class=\"c5b541d88d911dcbb84397def1a3492d\" data-index=\"1\" style=\"float: none; margin:0px 0 0px 0; text-align:center;\">\n<script async src=\"https:\/\/pagead2.googlesyndication.com\/pagead\/js\/adsbygoogle.js\"><\/script>\r\n<!-- m -->\r\n<ins class=\"adsbygoogle\"\r\n     style=\"display:block\"\r\n     data-ad-client=\"ca-pub-3943282228430444\"\r\n     data-ad-slot=\"5751019935\"\r\n     data-ad-format=\"auto\"\r\n     data-full-width-responsive=\"true\"><\/ins>\r\n<script>\r\n     (adsbygoogle = window.adsbygoogle || []).push({});\r\n<\/script>\n<\/div>\n<p> [ad_1]\n<\/p>\n<div id=\"id_text\" itemprop=\"articleBody\">\n<p>Need more convincing that it will soon be impossible to tell whether a video of a person is real or fake? Enter Samsung&#8217;s new research, in which a neural network can turn a still image into a disturbingly convincing video.\u00a0<\/p>\n<p>Researchers at the Samsung AI center in Moscow have achieved this, <a href=\"https:\/\/www.vice.com\/en_us\/article\/qv7zkw\/create-fake-videos-of-faces-samsung-ai-labs-algorithm\" target=\"_blank\" rel=\"noopener noreferrer\">Motherboard<\/a> reported Thursday, by training a &#8220;deep convolutional network&#8221; on a large number of videos showing talking heads, allowing it to identify certain facial features, and then using that knowledge to animate an image.\u00a0<\/p>\n<div class=\"see-also\">\n<p class=\"see-also-link\">SEE ALSO: <a href=\"http:\/\/mashable.com\/article\/petswap-deepfakes?utm_cid=a-seealso\" target=\"_blank\" rel=\"noopener\">Deepfake your pet with a fun, not-at-all terrifying algorithm<\/a><\/p>\n<\/div>\n<p>The results, presented in a <a href=\"https:\/\/arxiv.org\/abs\/1905.08233\" target=\"_blank\" rel=\"noopener noreferrer\">paper<\/a> called &#8220;Few-Shot Adversarial Learning of Realistic Neural Talking Head Models,&#8221; are not as good as some of the <a href=\"https:\/\/mashable.com\/2018\/04\/17\/jordan-peele-fake-obama-psa\/\" data-mashable-story-id=\"5ce7fa962463a3175f00000e\" target=\"_blank\" rel=\"noopener\">deepfake videos<\/a> you&#8217;ve seen, but to create those, you need a large number of images of the person you&#8217;re trying to animate. The advantage of Samsung&#8217;s approach is that you can turn a single still image (though the fidelity of the resulting video increases with more images) into a video.\u00a0<\/p>\n<p>You can see some of the results of this research in the video, below. Using a single still image of Fyodor Dostoevsky, Salvador Dali, Albert Einstein, Marilyn Monroe and even Mona Lisa, the AI was able to create videos of them talking which are realistic enough \u2014 at moments \u2014 to appear to be actual footage.\u00a0<\/p>\n<!-- Quick Adsense WordPress Plugin: http:\/\/quickadsense.com\/ -->\n<div class=\"c5b541d88d911dcbb84397def1a3492d\" data-index=\"2\" style=\"float: none; margin:0px 0 0px 0; text-align:center;\">\n<script async src=\"https:\/\/pagead2.googlesyndication.com\/pagead\/js\/adsbygoogle.js\"><\/script>\r\n<!-- 2 -->\r\n<ins class=\"adsbygoogle\"\r\n     style=\"display:block\"\r\n     data-ad-client=\"ca-pub-3943282228430444\"\r\n     data-ad-slot=\"5511153269\"\r\n     data-ad-format=\"auto\"\r\n     data-full-width-responsive=\"true\"><\/ins>\r\n<script>\r\n     (adsbygoogle = window.adsbygoogle || []).push({});\r\n<\/script>\n<\/div>\n\n<div class=\"youtube-wrapper\">\n<p><iframe src=\"http:\/\/www.youtube.com\/embed\/p1b5aiTrGzY\" data-channel=\"Egor Zakharov\" data-template=\"embedded\"><\/iframe><\/p>\n<\/div>\n<p>None of these videos will fool an expert, or anyone looking close enough. But as we&#8217;ve seen in <a href=\"https:\/\/mashable.com\/article\/nvidia-ai-generated-human-portraits\/\" target=\"_blank\" rel=\"noopener\">previous research<\/a> on AI-based generated imagery, the results tend to vastly improve in a matter of years.\u00a0<\/p>\n<p>The implications of this research are chilling. Armed with this tool, one only needs a single photo of a person (which are today easily obtainable for most people) to create a video of them talking. Add to that <a href=\"https:\/\/mashable.com\/2017\/07\/17\/scary-fake-video-tool-obama-say-anything\/\" data-mashable-story-id=\"5ce7fa962463a3175f00000e\" target=\"_blank\" rel=\"noopener\">a tool<\/a> that can use short snippets of sample audio material to create convincing, fake voice of a person, and one can get anyone to &#8220;say&#8221; anything. And with tools like <a href=\"https:\/\/mashable.com\/article\/nvidia-ai-generated-human-portraits\/\" target=\"_blank\" rel=\"noopener\">Nvidia&#8217;s GAN<\/a>, one could even create a realistic-looking, fake setting for such a video. As these tools become more powerful and easier to obtain, it will become tougher to tell real videos from fake ones; hopefully, the tools to discern between the two will get more advanced as well.\u00a0<\/p>\n<div class=\"bonus-video-card\">\n<h2 class=\"title\"><a href=\"https:\/\/mashable.com\/video\/google-nsynth-artificial-intelligence-synthesizer?utm_cid=a-bonusvideo\" target=\"_blank\" rel=\"noopener noreferrer\">WATCH: Google created an AI-based, open source music synthesizer<\/a><\/h2>\n<div class=\"jw-wrapper\">\n<p><iframe loading=\"lazy\" allowfullscreen=\"\" frameborder=\"0\" height=\"360\" src=\"https:\/\/mashable.com\/videos\/embed?video=dbl3agg203&amp;player=default\" width=\"640\"><\/iframe><\/p>\n<\/div>\n<\/div>\n<\/div>\n\n<!-- Quick Adsense WordPress Plugin: http:\/\/quickadsense.com\/ -->\n<div class=\"c5b541d88d911dcbb84397def1a3492d\" data-index=\"5\" style=\"float: none; margin:0px 0 0px 0; text-align:center;\">\n<script async src=\"https:\/\/pagead2.googlesyndication.com\/pagead\/js\/adsbygoogle.js\"><\/script>\r\n<!-- 5 -->\r\n<ins class=\"adsbygoogle\"\r\n     style=\"display:block\"\r\n     data-ad-client=\"ca-pub-3943282228430444\"\r\n     data-ad-slot=\"9258826580\"\r\n     data-ad-format=\"auto\"\r\n     data-full-width-responsive=\"true\"><\/ins>\r\n<script>\r\n     (adsbygoogle = window.adsbygoogle || []).push({});\r\n<\/script>\n<\/div>\n[ad_2]\n<br \/><a href=\"https:\/\/me.mashable.com\/tech\/5154\/watch-samsungs-new-ai-turn-mona-lisa-into-a-realistic-talking-head\" target=\"_blank\" rel=\"noopener\">Source link <\/a><\/p>\n\n<!-- Quick Adsense WordPress Plugin: http:\/\/quickadsense.com\/ -->\n<div class=\"c5b541d88d911dcbb84397def1a3492d\" data-index=\"3\" style=\"float: none; margin:0px 0 0px 0; text-align:center;\">\n<script async src=\"https:\/\/pagead2.googlesyndication.com\/pagead\/js\/adsbygoogle.js\"><\/script>\r\n<!-- 3 -->\r\n<ins class=\"adsbygoogle\"\r\n     style=\"display:block\"\r\n     data-ad-client=\"ca-pub-3943282228430444\"\r\n     data-ad-slot=\"2884989920\"\r\n     data-ad-format=\"auto\"\r\n     data-full-width-responsive=\"true\"><\/ins>\r\n<script>\r\n     (adsbygoogle = window.adsbygoogle || []).push({});\r\n<\/script>\n<\/div>\n\n<div style=\"font-size: 0px; height: 0px; line-height: 0px; margin: 0; padding: 0; clear: both;\"><\/div>","protected":false},"excerpt":{"rendered":"<p>[ad_1] Need more convincing that it will soon be impossible to tell whether a video of a person is real or fake? Enter Samsung&#8217;s new research, in which a neural network can turn a still image into a disturbingly convincing video.\u00a0 Researchers at the Samsung AI center in Moscow have achieved this, Motherboard reported Thursday, &hellip;<\/p>\n","protected":false},"author":1,"featured_media":11563,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[9],"tags":[],"_links":{"self":[{"href":"https:\/\/newsforher.com\/index.php?rest_route=\/wp\/v2\/posts\/11562"}],"collection":[{"href":"https:\/\/newsforher.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/newsforher.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/newsforher.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/newsforher.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=11562"}],"version-history":[{"count":1,"href":"https:\/\/newsforher.com\/index.php?rest_route=\/wp\/v2\/posts\/11562\/revisions"}],"predecessor-version":[{"id":11564,"href":"https:\/\/newsforher.com\/index.php?rest_route=\/wp\/v2\/posts\/11562\/revisions\/11564"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/newsforher.com\/index.php?rest_route=\/wp\/v2\/media\/11563"}],"wp:attachment":[{"href":"https:\/\/newsforher.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=11562"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/newsforher.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=11562"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/newsforher.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=11562"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}