AI-assisted 3D Preservation and Reconstruction of Temple Arts

Abstract

How does AI connect to the past in conservation? What can 17 years old photos be helpful in a renewed effort of preservation? This research aims to use AI to connect both in a seamless 3D reconstruction of heritage from imagery data taken from Gongfan Palace, Yunlin Taiwan. AI-assisted 3D modeling was used to reconstruct correspondent details across different 3D platforms of 3DGS or NeRF models generated by Postshot or KIRI Engine. Polygon or point models by Zephyr were referred to and assessed in two sets. The results also include AI-assist modeling outcomes in Stable Diffusion and Postshot-based animation. The evolved documenta-tion and interpretation in AI presents a novel arrangement of working processes contributed by new structure and management of resources, formats, and interfaces, as a continuous preservation effort.