OpenAI, which is an artificial intelligence research company and is responsible for chatbot, ChatGPT as well as text-to-image generator DALL-E has now come up with a new tool. This new tool is called Point-E.
Reports suggest that this new tool is quite difficult to use but at the same time can produce spectacular 3D object results.
As per a research paper put out describing the tool, it will allow users to come up with 3D objects just by putting in simple text input. Technically, this process is done by pairing a text-to-image AI model with an image-to- 3D model.
In simple words, when users type in a work such as ‘Santa dog’, upon typing this, the text-to-image model will generate an image accordingly. After this, the 3D model will produce a 3D object based on that sample image.
As can be seen from the images, the results won’t be up to par, however, it will come with an advantage, which is that the images will come out very fast.
According to the research paper, in order to make use of this AI technique, users will need to have a computer with a single graphics card and the result will come out within 1-2 minutes.
Now, you might be getting confused as to what is the difference between this method and DALL-E as both are involved in creating images from text. Well, DALL-E comes with a neural network which creates images from text captions. The company also claims that the DALL E 2, which is the second generation of artificial intelligence, has the capability to create more original and realistic images and art from a text description with 4x greater resolution.
Other than this company, Google also comes with a similar tool called DreamFusion which can create images in a similar manner. However, it requires much more powerful hardware to come out with a result.
ALSO READ | ChatGPT decoded: Everything to know about OpenAI’s new chatbot Internet can’t seem to get enough of
