Immediately after Meta debuted its own text-to-video AI product, “Make a Video,” Google unveiled Imagen Video, another text-to-video AI service.A textual prompt will be used exclusively by the tool to create films at 24 frames per second and 1280×768 resolution. Users of this beta tool can make Pixar-style monsters and fantastical architecture.
The tool from Google has been demonstrated with coffee being poured into a cup.The features of Imagen Video include animated letters displayed in various animation styles, 3D spinning objects, and created movies of works of art by well-known artists. The tool would also have internal data sets of 14 million films and 60 million photos, as well as noise cancellation augmentation.
A 16-frame movie will be produced once users input text instructions that have been encoded as textual embedding. The video would then be utilised to produce a 5.3s, 128-frame high definition video.
Google’s Imagen Video has drawn criticism for violating copyright laws, misusing artwork, and endangering upcoming artists while other businesses including Meta, Microsoft, and DALL-E2 have worked on comparable video creators.