The American company OpenAI, which develops and licenses technologies based on machine learning, presented an innovative solution for the arts. The DALL-E AI model has been equipped with a new function for generating visual components.
The essence of the innovative feature is that now you can expand the boundaries of the intended image. Like other models based on AI, the new functionality is able to convert text into images. But on the human side, a clear description of the new visual effects is needed.
“Test” began with famous paintings. DALL-E tried to “imagine” being in Johannes Vermeer's famous painting Girl with a Pearl Earring. However, even the limited input data did not affect the fact that the AI created an artist-style background with imitation of the shadows and highlights of the original.
But the new functionality can go differently. The scenario can develop in a direction radically opposite to the style of the painter's work. An example of this is the plot of the post-apocalyptic Mona Lisa.
The testers were also interested in DALL-E's capabilities in relation to known images. Visual components of advertising, cinematography, social networks, company logos.
It remains to expect interesting results that will pleasantly surprise or possibly disappoint. After all, the visual component is still subjective.
