{"id":20052,"date":"2022-09-27T07:52:00","date_gmt":"2022-09-27T14:52:00","guid":{"rendered":"http:\/\/jnack.com\/blog\/?p=20052"},"modified":"2022-09-26T10:04:07","modified_gmt":"2022-09-26T17:04:07","slug":"nvidias-get3d-promises-text-to-model-generation","status":"publish","type":"post","link":"http:\/\/jnack.com\/blog\/2022\/09\/27\/nvidias-get3d-promises-text-to-model-generation\/","title":{"rendered":"NVIDIA&#8217;s GET3D promises text-to-model generation"},"content":{"rendered":"\n<p>Depending on how well it works, tech like this could be the greatest unlock in 3D creation the world has ever known.<\/p>\n\n\n\n<figure class=\"wp-block-embed is-type-video is-provider-youtube wp-block-embed-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"wp-block-embed__wrapper\">\n<iframe loading=\"lazy\" title=\"NVIDIA GET3D: AI Model to Populate Virtual Worlds with 3D Objects and Characters\" width=\"604\" height=\"340\" src=\"https:\/\/www.youtube.com\/embed\/KfcsdhGKb1U?feature=oembed\" frameborder=\"0\" allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture\" allowfullscreen><\/iframe>\n<\/div><\/figure>\n\n\n\n<p>The company <a href=\"https:\/\/blogs.nvidia.com\/blog\/2022\/09\/23\/3d-generative-ai-research-virtual-worlds\/\">blog post<\/a> features interesting, promising details:<\/p>\n\n\n\n<blockquote class=\"wp-block-quote\"><p>Though quicker than manual methods, prior 3D generative AI models were limited in the level of detail they could produce. Even recent inverse rendering methods can only generate 3D objects based on 2D images taken from various angles, requiring developers to build one 3D shape at a time.<\/p><p>GET3D can instead churn out some 20 shapes a second when running\u00a0<a href=\"https:\/\/blogs.nvidia.com\/blog\/2016\/08\/22\/difference-deep-learning-training-inference-ai\/\">inference<\/a>\u00a0on a single NVIDIA GPU \u2014 working like a\u00a0<a href=\"https:\/\/developer.nvidia.com\/blog\/synthesizing-high-resolution-images-with-stylegan2\/\">generative adversarial network<\/a>\u00a0for 2D images, while generating 3D objects. [&#8230;]<\/p><p>GET3D gets its name from its ability to\u00a0<strong>G<\/strong>enerate\u00a0<strong>E<\/strong>xplicit\u00a0<strong>T<\/strong>extured\u00a0<strong>3D\u00a0<\/strong>meshes \u2014 meaning that the shapes it creates are in the form of a triangle mesh, like a papier-m\u00e2ch\u00e9 model, covered with a textured material. This lets users easily import the objects into game engines, 3D modelers and film renderers \u2014 and edit them.<\/p><\/blockquote>\n\n\n\n<p>See also <a href=\"https:\/\/ajayj.com\/dreamfields\">Dream Fields<\/a> (mentioned previously) from Google:<\/p>\n\n\n\n<figure class=\"wp-block-embed is-type-video is-provider-youtube wp-block-embed-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"wp-block-embed__wrapper\">\n<iframe loading=\"lazy\" title=\"Zero-Shot Text-Guided Object Generation with Dream Fields\" width=\"604\" height=\"340\" src=\"https:\/\/www.youtube.com\/embed\/1Fke6w46tv4?feature=oembed\" frameborder=\"0\" allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture\" allowfullscreen><\/iframe>\n<\/div><\/figure>\n","protected":false},"excerpt":{"rendered":"<p>Depending on how well it works, tech like this could be the greatest unlock in 3D creation the world has ever known. The company blog post features interesting, promising details: Though quicker than manual methods, prior 3D generative AI models were limited in the level of detail they could produce. Even recent inverse rendering methods [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[18,66],"tags":[],"_links":{"self":[{"href":"http:\/\/jnack.com\/blog\/wp-json\/wp\/v2\/posts\/20052"}],"collection":[{"href":"http:\/\/jnack.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/jnack.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/jnack.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/jnack.com\/blog\/wp-json\/wp\/v2\/comments?post=20052"}],"version-history":[{"count":2,"href":"http:\/\/jnack.com\/blog\/wp-json\/wp\/v2\/posts\/20052\/revisions"}],"predecessor-version":[{"id":20055,"href":"http:\/\/jnack.com\/blog\/wp-json\/wp\/v2\/posts\/20052\/revisions\/20055"}],"wp:attachment":[{"href":"http:\/\/jnack.com\/blog\/wp-json\/wp\/v2\/media?parent=20052"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/jnack.com\/blog\/wp-json\/wp\/v2\/categories?post=20052"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/jnack.com\/blog\/wp-json\/wp\/v2\/tags?post=20052"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}