The artificial intelligence DALL-E 2, developed by OpenAI, has managed to come to the fore many times with its talent, although it has not been opened to everyone’s access yet. The artificial intelligence tool, which transforms what is depicted in writing into a visual in the most realistic way, has become a hot topic on social media today with a different ability.
Kevin Parry, who has access to the early access version of DALL-E, included new works that are enchanting in his posts on Twitter. What made the aforementioned visuals special from other visuals was not the conversion of written definitions into visuals this time, but the perfect harmonizing of different visuals in a single frame.
DALL-E expands the image according to the context and theme of the image:
According to the post by Parry, DALL-E automatically displays any image presented to it, as requested. expands in size. This enlargement occurs not as an increase in the size of the image, but as a continuation of the same image from scratch – just like the ‘complete the paragraph’ questions in exams.
Moreover, DALL-E does this not only from one image, but from two images. He can also achieve this by ensuring that these two images are compatible with each other. This feature can also be combined with DALL-E’s ability to convert text to visuals. As a result, images like this emerge:
703
