Google has a native image feature within the ‘Geminai’ app. Open AI has expanded its image consistency and use with the identical measures as the combination of image creation function into the chat GPT.
Google announced on Wednesday that it has launched a picture generation function in Geminai. This update will step by step be provided to greater than 45 language users worldwide from today.
Google has a separate image generation model called ‘Imagen’. Nevertheless, this update is a direct integration of image functions inside Geminai like a chat GPT.
This feature allows users to directly enter and modify not only the pictures created by the AI, but additionally personal photos uploaded from mobile phones or computers.
As well as, it is feasible to significantly improve the consistency of the production image in addition to providing the consequence that’s sophisticated and contextual than the prevailing AI image generator. Google explains that it’s ‘multi-step editing’ and that text and image might be integrated to offer users with richer and more contextual responses.
For instance, a user can create a wide range of appearance, reminiscent of uploading the photos he took and changing the hair color.
As well as, it is feasible to create illustrations at the identical time while creating kid’s fairy tales as text. It is usually useful for making a 4 -cut cartoon like a chat GPT.
Google has launched and tested the native image creation of ‘Geminai 2.0 Flash’ for experiments through ‘AI Studio’ last March. On the time, this feature was well -received for its excellent quality, but at the identical time it was controversial that it could remove the watermark of the image.
The answer also introduced the answer to the watermark problem. “The image created or edited through Geminai is inserted in an invisible watermark,” he said.
“We’re also experimenting with adding a visual watermark.”
By Park Chan, reporter cpark@aitimes.com
