{"id":19829,"date":"2022-08-26T09:05:25","date_gmt":"2022-08-26T16:05:25","guid":{"rendered":"http:\/\/jnack.com\/blog\/?p=19829"},"modified":"2022-08-26T09:06:01","modified_gmt":"2022-08-26T16:06:01","slug":"death-metal-furby","status":"publish","type":"post","link":"http:\/\/jnack.com\/blog\/2022\/08\/26\/death-metal-furby\/","title":{"rendered":"&#x1f918;Death Metal Furby!&#x1f918;"},"content":{"rendered":"\n<p>See, isn&#8217;t that a more seductive title than &#8220;<a href=\"https:\/\/arxiv.org\/abs\/2208.01618\">Personalizing Text-to-Image Generation using Textual Inversion<\/a>&#8220;? &#x1f60c; But the so-titled paper seems really important in helping generative models like DALL\u2022E to become much more precise. The team writes:<\/p>\n\n\n\n<blockquote class=\"wp-block-quote\"><p>We ask: how can we use language-guided models to turn our cat into a painting, or imagine a new product based on our favorite toy? Here we present a simple approach that allows such creative freedom.<\/p><p>Using only 3-5 images of a user-provided concept, like an object or a style, we learn to represent it through new &#8220;words&#8221; in the embedding space of a frozen text-to-image model. These &#8220;words&#8221; can be composed into natural language sentences, guiding personalized creation in an intuitive way.<\/p><\/blockquote>\n\n\n\n<p>Check out the kind of thing it yields:<\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/arxiv.org\/abs\/2208.01618\"><img decoding=\"async\" loading=\"lazy\" width=\"579\" height=\"1008\" src=\"http:\/\/jnack.com\/blog\/wp-content\/uploads\/2022\/08\/Death-Metal-Furby.jpg\" alt=\"\" class=\"wp-image-19830\" srcset=\"http:\/\/jnack.com\/blog\/wp-content\/uploads\/2022\/08\/Death-Metal-Furby.jpg 579w, http:\/\/jnack.com\/blog\/wp-content\/uploads\/2022\/08\/Death-Metal-Furby-172x300.jpg 172w\" sizes=\"(max-width: 579px) 100vw, 579px\" \/><\/a><\/figure>\n","protected":false},"excerpt":{"rendered":"<p>See, isn&#8217;t that a more seductive title than &#8220;Personalizing Text-to-Image Generation using Textual Inversion&#8220;? &#x1f60c; But the so-titled paper seems really important in helping generative models like DALL\u2022E to become much more precise. The team writes: We ask: how can we use language-guided models to turn our cat into a painting, or imagine a new [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[66,2],"tags":[],"_links":{"self":[{"href":"http:\/\/jnack.com\/blog\/wp-json\/wp\/v2\/posts\/19829"}],"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=19829"}],"version-history":[{"count":5,"href":"http:\/\/jnack.com\/blog\/wp-json\/wp\/v2\/posts\/19829\/revisions"}],"predecessor-version":[{"id":19864,"href":"http:\/\/jnack.com\/blog\/wp-json\/wp\/v2\/posts\/19829\/revisions\/19864"}],"wp:attachment":[{"href":"http:\/\/jnack.com\/blog\/wp-json\/wp\/v2\/media?parent=19829"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/jnack.com\/blog\/wp-json\/wp\/v2\/categories?post=19829"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/jnack.com\/blog\/wp-json\/wp\/v2\/tags?post=19829"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}