Google recently updated its AI ecosystem to be capable of generating images from text, wherein users can generate AI images easily from Bard, and Google’s dedicated AI-image-generator platform ImageFX.
AI-image generators have become quite popular in recent days, reasoning to its proficiency of creating ultra-real images that even gets difficult to differentiate from the original. Few players including midjourney, Stable diffusion and DallE-2 have made that possible and now Google joins the club.
With the AI model “Imagen” launched last year, Google now makes it available across its platforms including Bard, Duet AI, SGE and also in its new standalone website “ImageFX”.
Imagen 2 by Google
The text-to-image generators of Google are powered by Imagen 2, the firm’s advanced AI model for AI image generation. Developed by Google’s DeepMind, Imagen 2 is capable of generating high quality images, with precise rendering tasks such as human faces and hands realistically, which most of the models lag behind.
Google Bard will get the capability of generating images for the first time, as Imagen 2 will power the AI. You can ask Bard to generate a photo by explaining it in a conversational prompt and for which, Bard can generate a high-quality image. When I asked Bard “Can you create me an image of a boy sitting on a beach shore?”, it gave me this. The feature is already available in most countries, but currently only in English.
Not only Bard, but Google Ads, Duet AI in workspace, Search Generative Experience (SGE) and Google Cloud’s Vertex AI will also be incorporated with this Imagen 2. So, if you’re working on Google platforms, you can access the text-to-image generating AI model there itself.
ImageFX
ImageFX is a specially designed platform dedicated wholly to the text-to-image generating experiments. ImageFX lets you create images with a simple text prompt, then easily modify them with a new take on prompting using expressive chips. Something that lacks when you do so with Bard. In Bard or in other platforms like SGE, Duet AI, you can’t edit or improvise the AI-generated pictures. But ImageFX lets you do that.
With ImageFX, Google tries to replicate midjourney in its own style. ImageFX is available in AI Test kitchen of Google.
Safety regarding AI images
As this artificial intelligence generated images pose a threat of disinformation and problematic content, Google implements a new method to identify AI images. “The images generated with Imagen 2 will be marked by SynthID, a tool developed by Google DeepMind that adds a digital watermark directly into the pixels of images generated by AI,” Google writes. These watermarks though are imperceptible to the human eye, can be detectable for identification.
People can also tap on “About this image” in any pictures available in Google, to get to know if the image is AI generated or not.
FAQ
Can Google Bard generate images?
Yes. Google Bard is now capable of generating AI images, as it got incorporated with Imagen 2 AI model.
What is the name of AI text-to-image generation website of Google?
ImageFX by Google is wholly dedicated to create innovative images from text prompts.
How will you differentiate AI-generated image with an original image?
Tap “About this image” on any image that you find in Google. If the image is AI-generated, it’ll be mentioned there as “AI-generated”.
(For more such interesting informational, technology and innovation stuffs, keep reading The Inner Detail).
Kindly add ‘The Inner Detail’ to your Google News Feed by following us!
