Every so often OpenAI brings a new creation into the world that people tend to marvel at. The latest example of OpenAI’s creativity is DALL-E, which is basically an AI that uses the same protocols as GPT-3 except this time these protocols would be used to create any kind of an image based on a description that is given. It’s truly remarkable to see the diversity of images that can be created, from images of animals and plants to articles of clothing and you can mix all of these items up to create unique images of your own as well.
With all of that having been said and now out of the way, it is important to note that AI that was able to generate images from text based descriptions have been around for a really long time so this is nothing special in and of itself. However, it is the neural network that is used in DALL-E that makes it unique and far superior to previous iterations of the same kind of program because of the fact that this is the sort of thing that could potentially end up allowing more complex and intricate images to be created with several different layers within them.
The main thing that makes DALL-E so special is the fact that you don’t have to delve into its code and pathways to make it create an image. Rather, you would just need to describe the image which makes it quite similar to a human graphic designer or someone similar to this sort of role. After all, you’re not going to perform brain surgery on your designer to have them make an image for you. Rather, you would just tell them what you are looking for and they would make it, which is basically what DALL-E manages to do as well.
The origin of DALL-E’s name is also quite interesting. It is a combination of the loveable animated robot WALL-E and the famous surrealist painter Dali, thereby combining two aspects of the art form that are both really important in their own way.
Read next: Researchers believe that curators show bias while training AI models for algorithm-generated art
With all of that having been said and now out of the way, it is important to note that AI that was able to generate images from text based descriptions have been around for a really long time so this is nothing special in and of itself. However, it is the neural network that is used in DALL-E that makes it unique and far superior to previous iterations of the same kind of program because of the fact that this is the sort of thing that could potentially end up allowing more complex and intricate images to be created with several different layers within them.
The main thing that makes DALL-E so special is the fact that you don’t have to delve into its code and pathways to make it create an image. Rather, you would just need to describe the image which makes it quite similar to a human graphic designer or someone similar to this sort of role. After all, you’re not going to perform brain surgery on your designer to have them make an image for you. Rather, you would just tell them what you are looking for and they would make it, which is basically what DALL-E manages to do as well.
The origin of DALL-E’s name is also quite interesting. It is a combination of the loveable animated robot WALL-E and the famous surrealist painter Dali, thereby combining two aspects of the art form that are both really important in their own way.
Read next: Researchers believe that curators show bias while training AI models for algorithm-generated art
