Google has introduced a new AI tool that can generate graphics from text input. You can enter any descriptive language into the programme, and it will generate an image as exactly as possible.
This isn’t the first time an image-creating AI model has gotten a lot of attention. DALL-E from OpenAI was also capable of creating graphics from text, but Google’s version is more realistic.
Imagen is a Google product developed by the Brain Team at Google Research. The search engine behemoth promises to be able to provide “new levels of photorealism and a deep level of language understanding.”
Google developed a benchmark called Drawbench to compare Imagen to other text-to-image algorithms. This benchmark has a set of 200 text prompts that were entered into several AI systems for image creation. These photos were then shown to human raters. According to Google, they “prefer Imagen over other models in side-by-side comparisons, both in terms of sample quality and image-text alignment.”
However, it is worth noting that the photographs displayed on Imagen’s website are chosen and may be the finest images produced by the programme. Not all of the pictures generated by the model will be correct.
Imagen, like other text-to-AI solutions, is not yet publicly available. Google believes it is not yet acceptable for the broader public and requires more refinement. In the future, the team may make it available to the public, although access will most likely be restricted.
You may now sample a restricted version of Imagen on their website. It simply allows you to write a description using pre-selected terms.
To read our blog on “‘Google is introducing new software supply chain & zero trust security services,” click here.
