Abstract:In this article, we consider a version of the challenging problem of learning from datasets whose size is too limited to allow generalisation beyond the training set. To address the challenge, we propose to use a transfer learning approach whereby the model is first trained on a synthetic dataset replicating features of the original objects. In this study, the objects were smartphone photographs of near-complete Roman terra sigillata pottery vessels from the collection of the Museum of London. Taking the repli… Show more
Rapid progress in digitisation and computer techniques have enabled noteworthy new pottery analysis applications in recent decades. We focus on analytical techniques directed specifically at archaeological pottery research in this survey and review the specific benefits these have brought in the field. We consider techniques based on heterogeneous sources such as drawings, photographs, 3D scans and CT volume data. The various approaches and methods are structured according to the main steps in pottery processing in archaeology: documentation, classification and retrieval. Within these categories we review the most relevant papers and identify their advantages and limitations. We evaluate both freely and commercially available analysis tools and databases. Finally, we discuss open problems and future challenges in the field of pottery analysis.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.