DALL-E, from OpenAI, is not the only artificial intelligence capable of generating images through a short text description. A few weeks ago Google also presented ‘Image’, an alternative to the AI of the company founded by Elon Musk (among others) which, according to the Mountain View firm itself, is capable of creating much more realistic and higher quality designs. Now, Microsoft has joined the competition. does it with NUWA-Infinity an AI that is not only capable of producing images from text, but also to convert a static design into a video.
Microsoft describes NUWA as “a multimodal generative model designed to generate high-quality images and video from given text, image, or video input.” Its operation, therefore, does not differ much from what DALL-E or even Image (Google) can do. However, it has a number of advantages over both models of artificial intelligence. It is the only AI that can generate a video from an image created using a text description. The AI also you can also generate video directly through a description.
“Compared to DALL-E, Image and Parti, NUWA-Infinity can generate high-resolution images with arbitrary sizes and also support long-time video generation.”
NUWA, Microsoft AI can also extend any type of image
NUWA, the Microsoft AI that generates images and videos from a Text description, is also capable of “extend” any image and create a larger, higher resolution image.n. Artificial intelligence, specifically, detects the information shown in the original photograph and, based on its parameters, generates a much more complete one. NUWA, for example, can extend ‘The Starry Night’, the work of Vincent van Gogh. It also does so with identical detail to that presented in the original design and a very precise sequel.
Microsoft, for the moment, has not offered more details about NUWA, beyond some examples that show the potential of this AI and how it is capable of converting a text into an image, an image into a video, or a text into a video, as well as the possibility of expanding any design. It is, without a doubt, an interesting option to DALL-E and Image, although both algorithms also have their advantages.
Picture eg generates much more realistic designs, although it is not yet available to users. DALL-E, on the other hand, offers less realistic images, but it is more accessible to users, as it is available through a public beta – although with limited access.