{"id":1156,"date":"2018-10-12T16:45:12","date_gmt":"2018-10-12T16:45:12","guid":{"rendered":"https:\/\/www.aiproblog.com\/index.php\/2018\/10\/12\/shadowplay-using-our-hands-to-have-some-fun-with-ai\/"},"modified":"2018-10-12T16:45:12","modified_gmt":"2018-10-12T16:45:12","slug":"shadowplay-using-our-hands-to-have-some-fun-with-ai","status":"publish","type":"post","link":"https:\/\/www.aiproblog.com\/index.php\/2018\/10\/12\/shadowplay-using-our-hands-to-have-some-fun-with-ai\/","title":{"rendered":"ShadowPlay: Using our hands to have some fun with AI"},"content":{"rendered":"<p>Author: <\/p>\n<div>\n<div class=\"block-paragraph\">\n<div class=\"rich-text\">\n<p><i><b>Editor\u2019s note:<\/b><a href=\"http:\/\/tensorflow.org\/\">TensorFlow<\/a>, our open source machine learning platform, is just that\u2014open to anyone. Companies, nonprofits, researchers and developers have used TensorFlow in some pretty cool ways and at Google, we&#8217;re always looking to do the same. Here&#8217;s one of those stories.<\/i><\/p>\n<\/div>\n<\/div>\n<div class=\"block-paragraph\">\n<div class=\"rich-text\">\n<p><a href=\"https:\/\/ich.unesco.org\/en\/RL\/chinese-shadow-puppetry-00421\">Chinese shadow puppetry<\/a>\u2014which uses silhouette figures and music to tell a story\u2014is an ancient Chinese art form that\u2019s been used by generations to charm communities and pass along cultural history. At Google, we\u2019re always experimenting with how we can <a href=\"https:\/\/blog.google\/outreach-initiatives\/arts-culture\/machine-learning-meets-culture\/\">connect culture with AI<\/a> and make it fun, which got us thinking: can AI help put on a shadow puppet show? <\/p>\n<p>So we created ShadowPlay, an interactive installation that celebrates the shadow puppetry art form. The installation, built using TensorFlow and TPUs, uses AI to recognize a person\u2019s hand gestures and then magically transform the shadow figure into digital animations representing the 12 animals of the Chinese zodiac and in an interactive show. <\/p>\n<\/div>\n<\/div>\n<div class=\"block-image_full_width\">\n<div class=\"article-module h-c-page\">\n<div class=\"h-c-grid\">\n<figure class=\"article-image--medium h-c-grid__col h-c-grid__col--4 h-c-grid__col--offset-4 \"><img decoding=\"async\" alt=\"Shadowplay.gif\" src=\"https:\/\/storage.googleapis.com\/gweb-uniblog-publish-prod\/original_images\/Shadowplay.gif\"><figcaption class=\"article-image__caption \">\n<div class=\"rich-text\">\n<p>Attendees use their hands to make shadow figures, which transform into animated characters and creates. <\/p>\n<\/div>\n<\/figcaption><\/figure>\n<\/div>\n<\/div>\n<\/div>\n<div class=\"block-paragraph\">\n<div class=\"rich-text\">\n<p>We debuted ShadowPlay at the <a href=\"http:\/\/www.waic2018.com\/\">World AI Conference<\/a> and <a href=\"https:\/\/www.google.cn\/intl\/en\/events\/developerdays2018\/\">Google Developers Day<\/a> in Shanghai in September. To build the experience, we developed a custom machine learning model that was trained on a dataset made up of lots of examples of people\u2019s hand shadows, which could eventually recognize the shadow and match it to the corresponding animal. \u201cIn order to bring this project to life, we asked Googlers to help us train the model by making a lot of fun hand gestures. Once we saw the reaction of users seeing their hand shadows morph into characters, it was impossible not to smile!\u201d, says Miguel de Andres-Clavera, Project Lead at Google. To make sure the experience could guess what animal people were making with high accuracy, we trained the model using TPUs, our custom machine learning hardware accelerators. <\/p>\n<\/div>\n<\/div>\n<div class=\"block-image_carousel\">\n<div class=\"h-c-page article-module\">\n<div class=\"article-module glue-pagination h-c-carousel h-c-carousel--simple h-c-carousel--dark ng-cloak\" data-glue-pagination-config=\"{cyclical: true}\">\n<div class=\"h-c-carousel__wrap\">\n<ul class=\"glue-carousel ng-cloak\" data-glue-carousel-options=\"{pointerTypes: ['touch', 'mouse'], jump: true}\">\n<li class=\"h-c-carousel__item article-carousel__slide\">\n<figure>\n<div class=\"article-carousel__slide-img\" style=\"background-image: url(https:\/\/storage.googleapis.com\/gweb-uniblog-publish-prod\/images\/Shadowplay_AI_3.max-2000x2000.jpg);\"><span class=\"h-u-visually-hidden\">Shadowplay_AI_3.jpg<\/span><\/div><figcaption class=\"article-carousel__caption h-c-copy h-u-mt-std\">\n<div class=\"rich-text\">\n<p>An attendee uses her hand to make a shadow figure, which the machine learning system recognizes as a dragon.<\/p>\n<\/div>\n<\/figcaption><\/figure>\n<\/li>\n<li class=\"h-c-carousel__item article-carousel__slide\">\n<figure>\n<div class=\"article-carousel__slide-img\" style=\"background-image: url(https:\/\/storage.googleapis.com\/gweb-uniblog-publish-prod\/images\/Shadowplay_AI_2.max-2000x2000.jpg);\"><span class=\"h-u-visually-hidden\">Shadowplay_AI_2.jpg<\/span><\/div><figcaption class=\"article-carousel__caption h-c-copy h-u-mt-std\">\n<div class=\"rich-text\">\n<p>A user can make shadow puppets from the Chinese zodiac, like this tiger, which the AI will recognize and turn it into an animated figure. <\/p>\n<\/div>\n<\/figcaption><\/figure>\n<\/li>\n<li class=\"h-c-carousel__item article-carousel__slide\">\n<figure>\n<div class=\"article-carousel__slide-img\" style=\"background-image: url(https:\/\/storage.googleapis.com\/gweb-uniblog-publish-prod\/images\/Shadowplay_AI_4.max-2000x2000.jpg);\"><span class=\"h-u-visually-hidden\">Shadowplay_AI_4.jpg<\/span><\/div><figcaption class=\"article-carousel__caption h-c-copy h-u-mt-std\">\n<div class=\"rich-text\">\n<p>An attendee\u2019s shadow figure, recognized by the machine learning system, transforms into an animated silhouette <\/p>\n<\/div>\n<\/figcaption><\/figure>\n<\/li>\n<li class=\"h-c-carousel__item article-carousel__slide\">\n<figure>\n<div class=\"article-carousel__slide-img\" style=\"background-image: url(https:\/\/storage.googleapis.com\/gweb-uniblog-publish-prod\/original_images\/Shadowplay_AI.gif);\"><span class=\"h-u-visually-hidden\">Shadowplay_AI.gif<\/span><\/div><figcaption class=\"article-carousel__caption h-c-copy h-u-mt-std\">\n<div class=\"rich-text\">\n<p>The ShadowPlay installation brings to life Chinese shadow puppetry with AI<\/p>\n<\/div>\n<\/figcaption><\/figure>\n<\/li>\n<\/ul>\n<div class=\"h-c-carousel__paginate glue-pagination-previous\" data-glue-pagination-label=\"Previous\" data-glue-pagination-update-model=\"false\">\n<div class=\"h-c-carousel__paginate-wrap\"><svg class=\"h-c-icon h-c-icon--keyboard-arrow-left\" role=\"img\"><use xlink:href=\"#mi-keyboard-arrow-right\"><\/use><\/svg><\/div>\n<\/div>\n<div class=\"h-c-carousel__paginate glue-pagination-next\" data-glue-pagination-label=\"Next\" data-glue-pagination-update-model=\"false\">\n<div class=\"h-c-carousel__paginate-wrap\"><svg class=\"h-c-icon h-c-icon--keyboard-arrow-right\" role=\"img\"><use xlink:href=\"#mi-keyboard-arrow-right\"><\/use><\/svg><\/div>\n<\/div>\n<\/div>\n<div class=\"h-c-carousel__navigation\">\n<div class=\"glue-pagination-page-list\"><\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<div class=\"block-paragraph\">\n<div class=\"rich-text\">\n<p>We had so much fun building ShadowPlay (almost as much fun as practicing our shadow puppets \u2026 ), that we\u2019ll be bringing it to more events around the world soon! <\/p>\n<\/div>\n<\/div>\n<\/div>\n<p><?php comment_form(); ?><br \/>\n<a href=\"https:\/\/www.blog.google\/technology\/ai\/shadowplay-using-our-hands-have-some-fun-ai\/\">Go to Source<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Author: Editor\u2019s note:TensorFlow, our open source machine learning platform, is just that\u2014open to anyone. Companies, nonprofits, researchers and developers have used TensorFlow in some pretty [&hellip;] <span class=\"read-more-link\"><a class=\"read-more\" href=\"https:\/\/www.aiproblog.com\/index.php\/2018\/10\/12\/shadowplay-using-our-hands-to-have-some-fun-with-ai\/\">Read More<\/a><\/span><\/p>\n","protected":false},"author":1,"featured_media":1157,"comment_status":"registered_only","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_bbp_topic_count":0,"_bbp_reply_count":0,"_bbp_total_topic_count":0,"_bbp_total_reply_count":0,"_bbp_voice_count":0,"_bbp_anonymous_reply_count":0,"_bbp_topic_count_hidden":0,"_bbp_reply_count_hidden":0,"_bbp_forum_subforum_count":0,"footnotes":""},"categories":[24],"tags":[],"_links":{"self":[{"href":"https:\/\/www.aiproblog.com\/index.php\/wp-json\/wp\/v2\/posts\/1156"}],"collection":[{"href":"https:\/\/www.aiproblog.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.aiproblog.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.aiproblog.com\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.aiproblog.com\/index.php\/wp-json\/wp\/v2\/comments?post=1156"}],"version-history":[{"count":0,"href":"https:\/\/www.aiproblog.com\/index.php\/wp-json\/wp\/v2\/posts\/1156\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.aiproblog.com\/index.php\/wp-json\/wp\/v2\/media\/1157"}],"wp:attachment":[{"href":"https:\/\/www.aiproblog.com\/index.php\/wp-json\/wp\/v2\/media?parent=1156"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.aiproblog.com\/index.php\/wp-json\/wp\/v2\/categories?post=1156"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.aiproblog.com\/index.php\/wp-json\/wp\/v2\/tags?post=1156"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}