Reconstructing Artificial Artefacts
AI Camera Simulation & Splatting Visualization
Research By
Jessica In (The Bartlett, UCL)
Exhibition
PhD Research Projects 2026, UCL
Role
Visual Synthesis & Technical Artist
As part of the doctoral research by Jessica In, this project explores the agency of generative AI in rewriting spatial memories. I was invited to translate static Gaussian Splatting data into a dynamic narrative using Unity-based AI camera simulations, effectively re-constructing "artificial artefacts" through a lens of digital archaeology.
The visualization employs an autonomous camera movement system that mimics human observation patterns within fragmented point-cloud environments.
By manipulating the "Gaussian Splatting" density and cinematic trajectories, we examine how machine learning models reinterpret architectural representations.
By manipulating the "Gaussian Splatting" density and cinematic trajectories, we examine how machine learning models reinterpret architectural representations.
This synthesis of computer vision and cinematic simulation creates an "auto-ethnographic" space, where personal memories are rewritten through the rigour of algorithmic reconstruction.