Accuracy and ambiguity Geometric reconstruction of a seventh century stone temple in Hanchey, Cambodia

Abstract

Modeling the fragmented and heavily eroded remains of early temple architecture poses several challenges in accurate reconstruction of shape and form from digitally acquired datasets. This paper describes a collection of stepwise ad-hoc modeling methods that can re-assemble ambiguous and fragmentary evidence to provide a robust and empirical platform for the reconstruction of ruined temples. The paper presents the results of the method and the degree of accuracy and ambiguity in the acquisition, processing and reconstruction phases. A key aspect of the method is the maintenance of multiple “ground truths” from plural sources of partial evidence. Key findings of the paper demonstrate early results from the manipulation of geometric modeling primitives based on point collections, an advance in extending the classical tools of architectural analysis and comparison. The problem of accuracy and ambiguity in these methods and their algorithmic implementation is the subject of further investigation.