Google Cloud has announced an upgrade to its Text-to-Image tool, Imagen 2, which has been developed from its previous version. This upgraded tool is now available for general use for Vertex AI customers on the waiting list.
Imagen 2 is developed using technology from Google DeepMind and offers several improvements compared to its previous version. Apart from better image quality, it also supports image generation for specific use cases such as creating product logos, applying logos as labels on products, or generating images that require text in different languages. Additionally, Imagen 2 also understands images and provides detailed descriptions of what it sees.
To ensure authenticity, Imagen 2’s images will have watermarks to confirm that they are not created by humans, but rather by DeepMind’s Synth ID system, which was introduced earlier.
Google has mentioned that Imagen API has been tested by companies like Snap, Shutterstock, and Canva.
TLDR: Google Cloud has released an upgraded version of its Text-to-Image tool, Imagen 2, which offers improved image quality and supports various use cases such as creating logos and generating images with text in different languages. Imagen 2 provides detailed descriptions of the images it generates and includes watermarks to ensure authenticity. Companies like Snap, Shutterstock, and Canva have already tested the Imagen API.
Leave a Comment