Cultural heritage is an integral element of today’s society. It is crucial to the creation of a common identity and the connection to people with similar backgrounds. Historical and heritage buildings are hard to study and organise in virtual platforms because of aging, destruction, and the geometric complexity of their structure. The ability to reason about the structure and the style of a heritage building is therefore of great importance to architects and cultural heritage experts, as it can assist them in its study and classification.

Different from previous work in the field of architecture which is confined to 2D based classification studies, ANNFASS proposes an innovative framework based on artificial neural networks for learning the structure and style of buildings from 3D data. The ANNFASS framework is composed of a neural network that can segment buildings into architectural components, a neural network that can detect the architectural style of a building and a shape grammar extraction and comparison method. The neural networks will be trained on an annotated building 3D dataset, and combined with the shape grammar to enable the ANNFASS software for: organizing buildings according to architectural style; observing the architectonic parts of a building and highlighting its main stylistic influences; analysing and comparing the design rules of architecture.

The objectives of this project are: to develop the neural networks and shape grammar; to collect annotated monument 3D data to train and test them; to develop and evaluate the software package for assisting architects and cultural heritage experts in their documentation workflows. The successful outcome of this project, ensured by the partners’ experience in computer graphics, geometry processing, machine learning and architecture, will have a definite impact on their scientific excellence and expertise, as well as society in general, through its application to the digital cultural heritage field.

CyI Principal Investigator

George Artopoulos
Email: This email address is being protected from spambots. You need JavaScript enabled to view it.
Tel. +357 22 208 619