Cubism – images generated by AI model – DALL-E

By admin | December, 30, 2022

DALL-E is a neural network-based image generation system developed by OpenAI. It is trained to generate images from textual descriptions, using a dataset of text-image pairs. Given a textual description as input, DALL-E is able to generate a corresponding image that visually represents the concepts described in the text.

DALL-E was inspired by the idea of a “conceptual art engine,” which is a system that can generate original artwork based on a set of predefined concepts. The name DALL-E is a reference to the artist Salvador Dali and the Pixar character WALL-E.

The way DALL-E works is similar to how other deep learning models work. It is trained on a large dataset of text-image pairs and learns to map text descriptions to images by minimizing the difference between the generated images and the ground truth images in the dataset.

During training, DALL-E uses an encoder-decoder architecture, with the encoder responsible for processing the input text and the decoder responsible for generating the output image. The encoder processes the input text through a series of layers, extracting relevant features and encoding them into a compact representation. The decoder then uses this representation to generate the output image, which is compared to the ground truth image and the difference is used to update the model’s parameters.

DALL-E is capable of generating a wide range of images based on the input text. It can generate images of animals, objects, scenes, and even abstract concepts.

It was not trained in a specific art style. However, the way in which it generates images is influenced by the training data that it was exposed to, so it is possible that the generated images may be influenced by the style of the images in the training dataset to some extent.

We asked ChatGPT if DALL-E can recreate Cubism and this is the answer we got:

“It is possible that DALL-E could be used to generate images that are inspired by the cubist art style, but it is unlikely that it could perfectly recreate the cubist style. Cubism is an art movement that was characterized by the use of geometric shapes and forms to depict objects and figures, often in a fragmented or abstracted way. It is a highly stylized art form that requires a great deal of artistic skill and creativity to master.

DALL-E is a machine learning model that is trained to generate images based on textual descriptions, rather than to recreate specific art styles. While it may be able to generate images that incorporate elements of the cubist style, such as the use of geometric shapes and forms, it is unlikely that it could perfectly replicate the complexity and nuance of the cubist style.

(…)  it is important to remember that it is a machine learning model and is not capable of replicating the full range of artistic styles and techniques that humans are capable of.

Its ability to generate detailed and realistic images has been quite impressive, and it has the potential to be used for a variety of applications in the field of artificial intelligence and machine learning.”

ChatGPT asked if DALL-E can generate Cubism images.


We decided give it a go and try it ourselves.
The result is not bad at all, see below for yourself!