{"id":81959,"date":"2022-09-29T17:43:01","date_gmt":"2022-09-29T17:43:01","guid":{"rendered":"https:\/\/harchi90.com\/meta-announces-make-a-video-which-generates-video-from-text\/"},"modified":"2022-09-29T17:43:01","modified_gmt":"2022-09-29T17:43:01","slug":"meta-announces-make-a-video-which-generates-video-from-text","status":"publish","type":"post","link":"https:\/\/harchi90.com\/meta-announces-make-a-video-which-generates-video-from-text\/","title":{"rendered":"Meta announces Make-A-Video, which generates video from text"},"content":{"rendered":"<div itemprop=\"articleBody\">\n<figure class=\"intro-image intro-left\">\n  <figcaption class=\"caption\">\n<div class=\"caption-text\">enlarge <span class=\"sep\">\/<\/span> Still image from an AI-generated video of a teddy bear painting a portrait.<\/div>\n<\/figcaption><\/figure>\n<aside id=\"social-left\" class=\"social-left\" aria-label=\"Read the comments or share this article\">\n<\/aside>\n<p><!-- cache hit 363:single\/related:5adae47189229f707027b5cea9c93fa7 --><!-- empty --><\/p>\n<p>Today, Meta announced Make-A-Video, an AI-powered video generator that can create novel video content from text or image prompts, similar to existing image synthesis tools like DALL-E and Stable Diffusion.  It can also make variations of existing videos, though it&#8217;s not yet available for public use.<\/p>\n<p>On Make-A-Video&#8217;s announcement page, Meta shows example videos generated from text, including &#8220;a young couple walking in heavy rain&#8221; and &#8220;a teddy bear painting a portrait.&#8221;  It also showcases Make-A-Video&#8217;s ability to take a static source image and animate it.  For example, a still photo of a sea turtle, once processed through the AI \u200b\u200bmodel, can appear to be swimming.<\/p>\n<p>The key technology behind Make-A-Video\u2014and why it has arrived sooner than <a rel=\"nofollow noopener\" href=\"https:\/\/twitter.com\/WriteArthur\/status\/1575487917828370436?s=20&amp;t=hq9KSyFCqtSSW__ImdeT1g\" target=\"_blank\">some experts<\/a> anticipated\u2014is that it builds off existing work with text-to-image synthesis used with image generators like OpenAI&#8217;s DALL-E.  In July, Meta announced its own text-to-image AI model called Make-A-Scene.<\/p>\n<p>Instead of training the Make-A-Video model on labeled video data (for example, captioned descriptions of the actions depicted), Meta instead took image synthesis data (still images trained with captions) and applied unlabeled video training data so the model learns a sense of where a text or image prompt might exist in time and space.  Then it can predict what comes after the image and display the scene in motion for a short period.<\/p>\n<aside class=\"ad_wrapper\" aria-label=\"In Content advertisement\">\n    <span class=\"ad_notice\">Advertisement <\/span>    <\/p>\n<\/aside>\n<div class=\"gallery shortcode-gallery gallery-wide\">\n<ul>\n<li data-thumb=\"https:\/\/cdn.arstechnica.net\/wp-content\/uploads\/2022\/09\/A_teddy_bear_painting_a_portrait-150x150.gif\" data-src=\"https:\/\/cdn.arstechnica.net\/wp-content\/uploads\/2022\/09\/A_teddy_bear_painting_a_portrait.gif\" data-responsive=\"https:\/\/cdn.arstechnica.net\/wp-content\/uploads\/2022\/09\/A_teddy_bear_painting_a_portrait.gif 1080, https:\/\/cdn.arstechnica.net\/wp-content\/uploads\/2022\/09\/A_teddy_bear_painting_a_portrait.gif 2560\" data-sub-html=\"#caption-1885549\">\n<figure style=\"height:512px;\"><figcaption id=\"caption-1885549\">\n                  <span class=\"icon caption-arrow icon-drop-indicator\"\/><\/p>\n<p>                      A video of a teddy bear painting a portrait, created with Meta&#8217;s Make-A-Video AI model (converted to GIF for display here).                    <\/p>\n<\/figcaption><\/figure>\n<\/li>\n<li data-thumb=\"https:\/\/cdn.arstechnica.net\/wp-content\/uploads\/2022\/09\/couple_in_rain-150x150.gif\" data-src=\"https:\/\/cdn.arstechnica.net\/wp-content\/uploads\/2022\/09\/couple_in_rain.gif\" data-responsive=\"https:\/\/cdn.arstechnica.net\/wp-content\/uploads\/2022\/09\/couple_in_rain.gif 1080, https:\/\/cdn.arstechnica.net\/wp-content\/uploads\/2022\/09\/couple_in_rain.gif 2560\" data-sub-html=\"#caption-1885547\">\n<figure style=\"height:512px;\"><figcaption id=\"caption-1885547\">\n                  <span class=\"icon caption-arrow icon-drop-indicator\"\/><\/p>\n<p>                      A video of &#8220;a young couple walking in a heavy rain&#8221; created with Make-A-Video.                    <\/p>\n<\/figcaption><\/figure>\n<\/li>\n<li data-thumb=\"https:\/\/cdn.arstechnica.net\/wp-content\/uploads\/2022\/09\/seaturtle-150x150.gif\" data-src=\"https:\/\/cdn.arstechnica.net\/wp-content\/uploads\/2022\/09\/seaturtle.gif\" data-responsive=\"https:\/\/cdn.arstechnica.net\/wp-content\/uploads\/2022\/09\/seaturtle.gif 1080, https:\/\/cdn.arstechnica.net\/wp-content\/uploads\/2022\/09\/seaturtle.gif 2560\" data-sub-html=\"#caption-1885548\">\n<figure style=\"height:512px;\"><figcaption id=\"caption-1885548\">\n                  <span class=\"icon caption-arrow icon-drop-indicator\"\/><\/p>\n<p>                      Video of a sea turtle, animated from a still image with Make-A-Video.                    <\/p>\n<\/figcaption><\/figure>\n<\/li>\n<\/ul><\/div>\n<p>&#8220;Using function-preserving transformations, we extend the spatial layers at the model initialization stage to include temporal information,&#8221; Meta wrote in a white paper.  &#8220;The extended spatial-temporal network includes new attention modules that learn temporal world dynamics from a collection of videos.&#8221;<\/p>\n<p>Meta has not made an announcement about how or when Make-A-Video might become available to the public or who would have access to it.  Meta provides a sign-up form people can fill out if they are interested in trying it in the future.<\/p>\n<p>Meta acknowledges that the ability to create photorealistic videos on demand presents certain social hazards.  At the bottom of the announcement page, Meta says that all AI-generated video content from Make-A-Video contains a watermark to &#8220;help ensure viewers know the video was generated with AI and is not a captured video.&#8221;<\/p>\n<p>If history is any guide, <a rel=\"nofollow noopener\" href=\"https:\/\/twitter.com\/EMostaque\/status\/1575499753445789697?s=20&amp;t=dxXL7L2jKZU6H5FA17hB8w\" target=\"_blank\">competitive<\/a> open source text-to-video models may follow (some, like CogVideo, already exist), which could make Meta&#8217;s watermark safeguard irrelevant.<\/p>\n<\/p><\/div>\n<p><script async src=\"\/\/platform.twitter.com\/widgets.js\" charset=\"utf-8\"><\/script><\/p>\n","protected":false},"excerpt":{"rendered":"<p>enlarge \/ Still image from an AI-generated video of a teddy bear painting a portrait. Today, Meta announced Make-A-Video, an AI-powered video generator that can create novel video content from text or image prompts, similar to existing image synthesis tools like DALL-E and Stable Diffusion. It can also make variations of existing videos, though it&#8217;s &hellip;<\/p>\n<p class=\"read-more\"> <a class=\"\" href=\"https:\/\/harchi90.com\/meta-announces-make-a-video-which-generates-video-from-text\/\"> <span class=\"screen-reader-text\">Meta announces Make-A-Video, which generates video from text<\/span> Read More &raquo;<\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"site-sidebar-layout":"default","site-content-layout":"default","ast-global-header-display":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","spay_email":"","jetpack_publicize_message":"","jetpack_is_tweetstorm":false,"jetpack_publicize_feature_enabled":true},"categories":[4],"tags":[],"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack-related-posts":[{"id":83682,"url":"https:\/\/harchi90.com\/meta-debuts-make-a-video-ai-system-that-generates-videos-from-nothing-but-text-prompts\/","url_meta":{"origin":81959,"position":0},"title":"Meta debuts &#8216;Make-A-Video&#8217; AI system that generates videos from nothing but TEXT prompts","date":"October 1, 2022","format":false,"excerpt":"Meta debuts artificial intelligence system that generates videos from nothing but TEXT prompts - with some creepy, surreal resultsMeta's new artificial intelligence can generate video clips from nothing but text The AI \u200b\u200bsystem, called Make-A-Video, was trained with combination of existing captioned images and unlabeled videos Thus far, it produces\u2026","rel":"","context":"In &quot;Technology&quot;","img":{"alt_text":"Meta debuted a new type of art-generating AI that can create videos from nothing but text prompts - with results that are creepy, surreal and impressive","src":"https:\/\/i0.wp.com\/i.dailymail.co.uk\/1s\/2022\/09\/30\/16\/62969991-11266897-Meta_debuted_a_new_type_of_art_generating_AI_that_can_create_vid-a-27_1664550127426.jpg?resize=350%2C200&ssl=1","width":350,"height":200},"classes":[]},{"id":16192,"url":"https:\/\/harchi90.com\/metas-latest-generative-ai-system-creates-stunning-images-from-sketches-and-text\/","url_meta":{"origin":81959,"position":1},"title":"Meta&#8217;s latest generative AI system creates stunning images from sketches and text","date":"July 16, 2022","format":false,"excerpt":"Meta Platforms Inc. today unveiled an advanced \u201cgenerative artificial intelligence system\u201d that's designed to help artists better showcase their creativity. The system, called \u201cMake-A-Scene,\u201d is meant to demonstrate how AI has the potential to empower anyone to bring their imagination to life. The user can simply describe and illustrate their\u2026","rel":"","context":"In &quot;Technology&quot;","img":{"alt_text":"","src":"https:\/\/i0.wp.com\/d15shllkswkct0.cloudfront.net\/wp-content\/blogs.dir\/1\/files\/2022\/07\/1-776x600.png?resize=350%2C200&ssl=1","width":350,"height":200},"classes":[]},{"id":36486,"url":"https:\/\/harchi90.com\/meta-is-putting-its-latest-ai-chatbot-on-the-web-for-the-public-to-talk-to\/","url_meta":{"origin":81959,"position":2},"title":"Meta is putting its latest AI chatbot on the web for the public to talk to","date":"August 5, 2022","format":false,"excerpt":"Meta's AI research labs have created a new state-of-the-art chatbot and are letting members of the public talk to the system in order to collect feedback on its capabilities. The bot is called BlenderBot 3 and can be accessed on the web. (Though, right now, it seems only residents in\u2026","rel":"","context":"In &quot;Technology&quot;","img":{"alt_text":"","src":"","width":0,"height":0},"classes":[]},{"id":83214,"url":"https:\/\/harchi90.com\/meta-is-using-ai-to-generate-videos-from-just-a-few-words\/","url_meta":{"origin":81959,"position":3},"title":"Meta is using AI to generate videos from just a few words","date":"September 30, 2022","format":false,"excerpt":"CNN Business \u2014 Artificial intelligence is getting better and better at generating an image in response to a handful of words, with publicly available AI image generators such as DALL-E 2 and Stable Diffusion. Now, Meta are researchers taking AI a step further: they're using it to concoct videos from\u2026","rel":"","context":"In &quot;Technology&quot;","img":{"alt_text":"","src":"","width":0,"height":0},"classes":[]},{"id":76310,"url":"https:\/\/harchi90.com\/meta-shows-stunning-full-body-tracking-only-via-quest-headset\/","url_meta":{"origin":81959,"position":4},"title":"Meta shows stunning full body tracking only via Quest headset","date":"September 24, 2022","format":false,"excerpt":"Image: meta Der Artikel kann nur mit aktiviertem JavaScript dargestellt werden. Bitte aktiviere JavaScript in deinem Browser und lade die Seite neu. So far, VR systems have tracked the head and hands. That could change soon: The predictive talent of artificial intelligence enables realistic full-body tracking and thus a better\u2026","rel":"","context":"In &quot;Technology&quot;","img":{"alt_text":"Meta shows stunning full body tracking only via Quest headset","src":"https:\/\/i0.wp.com\/mixed-news.com\/en\/wp-content\/uploads\/2022\/09\/QuestSim_-Human-Motion-Tracking-from-Sparse-Sensors-with-Simulated-Avatars-2-2-screenshot-375x211.png?resize=350%2C200&ssl=1","width":350,"height":200},"classes":[]},{"id":15167,"url":"https:\/\/harchi90.com\/metas-make-a-scene-ai-blends-human-and-computer-imagination-into-algorithmic-art\/","url_meta":{"origin":81959,"position":5},"title":"Meta&#8217;s &#8216;Make-A-Scene&#8217; AI blends human and computer imagination into algorithmic art","date":"July 15, 2022","format":false,"excerpt":"Text-to-image generation is the hot algorithmic process right now, with OpenAI's Craiyon (formerly DALL-E mini) and Google's Imagen AIs unleashing tidal waves of wonderfully weird procedurally generated art synthesized from human and computer imaginations. On Tuesday, Meta revealed that it too has developed an AI image generation engine, one that\u2026","rel":"","context":"In &quot;Technology&quot;","img":{"alt_text":"","src":"","width":0,"height":0},"classes":[]}],"fifu_image_url":"https:\/\/cdn.arstechnica.net\/wp-content\/uploads\/2022\/09\/makevideo_teddy_hero-760x380.jpg","_links":{"self":[{"href":"https:\/\/harchi90.com\/wp-json\/wp\/v2\/posts\/81959"}],"collection":[{"href":"https:\/\/harchi90.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/harchi90.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/harchi90.com\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/harchi90.com\/wp-json\/wp\/v2\/comments?post=81959"}],"version-history":[{"count":0,"href":"https:\/\/harchi90.com\/wp-json\/wp\/v2\/posts\/81959\/revisions"}],"wp:attachment":[{"href":"https:\/\/harchi90.com\/wp-json\/wp\/v2\/media?parent=81959"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/harchi90.com\/wp-json\/wp\/v2\/categories?post=81959"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/harchi90.com\/wp-json\/wp\/v2\/tags?post=81959"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}