Unsupervised Clustering of Roman Potsherds via Variational Autoencoders

TL;DR

This paper introduces an AI-based method using variational autoencoders for unsupervised clustering of Roman potsherds, aiding archaeologists in classification and analysis of fragmented pottery with a new database and visualization tool.

Contribution

It presents a novel deep learning approach for clustering archaeological pottery fragments, along with a comprehensive database and software tool for shape similarity analysis.

Findings

Successful clustering of potsherd profiles based on learned features

Creation of a large, annotated potsherd database (ROCOPOT)

Enhanced archaeological classification and research insights

Abstract

In this paper we propose an artificial intelligence imaging solution to support archaeologists in the classification task of Roman commonware potsherds. Usually, each potsherd is represented by its sectional profile as a two dimensional black-white image and printed in archaeological books related to specific archaeological excavations. The partiality and handcrafted variance of the fragments make their matching a challenging problem: we propose to pair similar profiles via the unsupervised hierarchical clustering of non-linear features learned in the latent space of a deep convolutional Variational Autoencoder (VAE) network. Our contribution also include the creation of a ROman COmmonware POTtery (ROCOPOT) database, with more than 4000 potsherds profiles extracted from 25 Roman pottery corpora, and a MATLAB GUI software for the easy inspection of shape similarities. Results are commented both from a mathematical and archaeological perspective so as to unlock new research directions in both communities.