Submitted by Franz Rosati


Location: Roma, Italy

Completion date: 2022

Project Team

Art Direction / Programming / Production / Live

Franz Rosati

3D / Surfacing Artist

Ruben Piergiovanni

Machine Learning Assistant (StyleGAN / SGAN)

Felice Colucci

Machine Learning Assistant (SampleRNN)

Simone Pelosi


Latentscape is a new media art project conceived to be exhibited as a series of digital artworks, a site-specific installation, a video art film or performed as an audiovisual concert.
The project depicts the exploration of virtual landscapes and territories.
The depicted immaterial dimension observes itself and begins to explore boundless territories, where streams of pixels take the form of cracks, erosions, mountains and colors, textures and sounds are the virtual representation of memories and emotions that the territory expresses through its shapes.
All elevation maps of landscapes and sounds are generated by different Machine Learning processes, known as GANs and Recursive Neural Networks, trained with personal datasets composed of publicly available satellite imagery and historical music collections, mixed and weighted with unrelated materials, to mislead the training and lead to unexpected results.
While the concepts of map and border appear as mere theoretical and functional superstructure, the landscapes and territories with their nuances, inlets, breaks, colors and topological diversity depict the meeting of human data in the border points, marking the landmarks of meeting and fusion of thoughts, memories and experiences.


the main goal for latentscape is to be exhibited in galleries, museums, as well as public projections and screenings. Since Latentscape is a multifaced project that can be presented as a live audiovisual concert as well as installation, it can have several different outputs adaptable for different setup and exhibitions.