Artificial Intelligence (AI) is an increasingly used resource. In the end, it tries to get computers to simulate human intelligence in a certain way. Their techniques are used when it is necessary to incorporate in a computer system, knowledge or characteristics of the human being to create models that allow to obtain simulate, in the most real way possible, the result obtained if it were done by a person.
CrAIyon, Artificial Intelligence using phrases
Craiyon, formerly known as DALL· E mini, is an AI model that can draw any image from any text message. It uses a model called “DALLE mini”, specifically an expanded version known as “DALLE mega” and is trained using Google TRC. More details can be found in W&B Project Report and DALL· E mini model card.
How to create more accurate images
It is very important to be precise in what you are looking for. There are some words that may be interesting to use to better define what we want to achieve. Words such as “illustration”, “photorealistic”, “high definition”, all in English, can be very useful to improve the result obtained.
The images obtained sometimes have a very strange appearance, especially the eyes, but it is due to a limitation in the coding of images and is expected to improve in future versions of the tool.
Example of using artificial intelligence
As you can put any text it is important to be very refined in the model we want to achieve. Here I show several examples, each more accurate. The first shows the result by putting the word cat. The page is programmed to recognize words written in English.
If we create a longer text and put black and white cat in a car this would be the result:
We can continue adding text that the result will be different but that takes into account what we write as in the following figure where rainy day has been added.
In addition, we can use different criteria, which the AI can recognize and create the image from what we put. In the same way you can change the result by adding a location obtaining what is seen in the following figure.
The result you get using this tool is really impressive. The advance in Generative-AI in the coming years will be exponential and will offer a new universe of possibilities that can help the growth of different branches such as mathematics, graphic design, drawing or illustrations of various kinds.
You can find more interesting articles about IA in https://avertigoland.com/category/technology/machine-learning/