Ways of Seeing: Exploring Ai Applications In Digital Reconstruction of Historic Building

This paper critically examines the potential of Artificial Intelligence (AI) tools in the digital reconstruction of lost or partially lost historical buildings for heritage recording and dissemination. Despite advancements in AR/VR applications, limitations persist in capturing the tectonic, tactile and spatial qualities of these buildings. Current AI technologies offer the potential to overcome these limitations. The study explores two scenarios: reconstructing the partially lost Buddhist Monastery of Sompur Mahavihara using textual evidence, and recreating the lost architecture of the Dennys Lascelles Wool store through existing visual records. These scenarios illustrate the complexities of digital reconstruction and the effectiveness of AI tools in capturing spatial qualities and tactile experiences. By comparing these scenarios, the paper highlights AI's potential to enhance historical research, provide immersive user experiences, and significantly contribute to heritage preservation and dissemination.