Generative Networks for Point Cloud Generation in Cultural Heritage Domain

Abstract

In the Cultural Heritage (CH) domain, the semantic segmentation of 3D point clouds with Deep Learning (DL) techniques allows to recognize historical architectural elements, at a suitable level of detail, and hence expedite the process of modelling historical buildings for the development of BIM models from survey data. However, it is more difficult to collect a balanced dataset of labelled architectural elements for training a network. In fact, the CH objects are unique, and it is challenging for the network to recognize this kind of data. In recent years, Generative Networks have proven to be proper for generating new data. Starting from such premises, in this paper Generative Networks have been used for augmenting a CH dataset. In particular, the performances of three state-of-art Generative Networks such as PointGrow, PointFLow and PointGMM have been compared in terms of Jensen-Shannon Divergence (JSD), the Minimum Matching Distance-Chamfer Distance (MMD-CD) and the Minimum Matching Distance-Earth Mover’s Distance (MMD-EMD). The objects generated have been used for augmenting two classes of ArCH dataset, which are columns and windows. Then a DGCNN-Mod network was trained and tested for the semantic segmentation task, comparing the performance in the case of the ArCH dataset without and with augmentation.

A Smart Heritage System to Re-Generate New Zealand’s 19th Century Timber Churches

“…Although smart, automated technologies have been applied by application of a procedural approach in our work, there is the opportunity to achieve more powerful outcomes using technologies such as Generative Neural Networks [25,26]. In a project that has similar motives to ours, Zheng and Yuan [27] have exploited machine learning technology in an urban design context using an artificial neural network that is first trained with generated design data, and then tested by adjusting the feature parameters.…”