about AI mosaics

The beauty of AI-generated art lies in its ability to explore new frontiers, challenge traditional notions of creativity, and offer fresh perspectives.

Three types of images are created using deep learning algorithms.
- The main image, such as a face, which is composed of
- hundreds of similar tiles
- Images of rooms, where the final images are presented as framed images.

Deep learning, text-to-image models are used to generate detailed images conditioned on text descriptions. In this project, the STABLE DIFFUSION model is used. It generates thousands of more or less similar images triggered by carefully chosen descriptive text input. The same text input is used for a set of images, and variations in image composition are made by intelligent parameter settings. For example, thousands of images of human faces can be generated.

These tiled images are used to create mosaics↗. The zoom player allows close inspection of the entire image composition.

The final images are 80x100 cm and can be printed on any suitable material, e.g. Alu DIBOND® .

To get an impression of what the final prints will look like, room images↗ are generated by the SD program. Finally, the mosaics are mounted in the room using Photoshop.

The project is not finished, new stuff will be added on an unpredictable base :).

room views

gallery of eight faces

in rooms

zooming

what is this about?

This WYSIWYG Web Builder template includes:
• Layout Grids
• Carousel
• Bootstrap Cards
• Responsive Menu

Lorem ipsum dolor sit amet, consectetur adipisicing elit. Corporis, omnis doloremque non cum id reprehenderit, quisquam totam aspernatur tempora minima unde aliquid ea culpa sunt. Reiciendis quia dolorum ducimus unde