Neural Terrain

Neural Terrain looks at how a machine can reinterpret and rebuild natural landscapes. I used pixel-sorting techniques to generate evolving terrains and topographic forms, letting computation echo natural patterns while also breaking them apart.

The black-and-white visuals at the beginning reflect the binary structure of computation. As the piece shifts into vivid color, the machine tries to imitate nature. This gradual change highlights the overlap and tension between the two, prompting viewers to think about how digital systems learn by observing, copying, and reshaping organic structures.

The work also considers how a machine “sees” and reorganizes physical space. By treating landscape as something that can be read, sorted, and rebuilt, the project raises questions about how we look to nature when building technology, and how machine vision, environmental simulation, and digital world-building influence that process.

Overall, the piece challenges the idea that nature and technology exist on opposite ends. Neural Terrain instead reveals how closely they are connected, and how digital terrains increasingly shape the way we understand our environment.