Google announces live image feature for Imagen 2.0 to enhance Vertex AI

At the Google Cloud Next event, the search giant announced lots of updates and new features for platforms to enhance Vertex AI. And of all the exciting details that came out from the event, one of the noteworthy announcements was updates to Imagen 2.0, Google’s image generation models.

Google has announced that Imagen 2.0 models can now be used to create short, 4-second live images from text prompts. Additionally, editing capabilities like inpainting/outpainting and digital watermarking are now generally available in Imagen 2.0.

Imagen 2.0’s advanced image generation and photo editing features

Text-to-live image capabilities are now available as a preview, making Imagen even more powerful on Vertex AI. According to Google’s official blog post, users can generate various kinds of four seconds live images such as GIFs with 24 frames per second (fps) and a resolution of 360×640 pixels. However, it won’t be that way forever, as the company has plans for “continuous enhancements” to make it even more powerful. Also noteworthy is the fact that live image generation is equipped with safety filters and digital watermarks.

Imagen 2.0 on Vertex AI also comes with advanced photo editing features. Some of the tweaks you can make to your images include removing unwanted elements in an image, adding new elements, and expanding the borders of the image to create a wider field of view. In addition, users can “generate invisible watermarks and verify images and live images generated by the Imagen family of models.” You can check out how some editing functionalities in Imagen 2.0 work in the official video below and let us know in the comments if you’re impressed with them.