Research Project Overview and Description
This project utilizes advanced AI and data science to revolutionize the study of ancient pottery from the Ararat Plain Southeast Archaeological Project in Armenia. By employing an AI-enabled robot arm for 3D scanning and developing machine learning algorithms for classification and digital reconstruction, the research aims to analyze fragmented artifacts at scale. The initiative seeks to uncover historical details about ancient vessel usage and site abandonment while advancing AI’s capability to process complex real-world datasets.
This project aims to revolutionize archaeological research by applying advanced Artificial Intelligence and data science to study ancient pottery from the Ararat Plain Southeast Archaeological Project in Armenia. Because pottery is often found in fragmented states, it is challenging to analyze at scale. To overcome this, the team will utilize an AI-enabled robot arm to efficiently digitize, photograph, weigh, and 3D scan these physical fragments. Furthermore, researchers will develop machine learning and computer vision algorithms to classify the pottery, analyze surface colors, and digitally puzzle fragments back into complete vessels using 3D models. This process will help uncover historical details about site abandonment and ancient vessel usage.
Research Outcome
The project will yield significant academic and practical deliverables. Initially, preliminary findings will form the foundation of a General Research Fund (GRF) application in Fall 2026, alongside a submission to a computer science conference. By 2027, the team will apply for the Innovation and Technology Support Programme (ITSP) to scale these methods for global use by archaeologists and museums. Furthermore, the research will produce articles for top-tier journals like Digital Applications in Archaeology and Cultural Heritage. Beyond academia, the project promises to revolutionize historical artifact analysis and advance AI algorithms by pushing their capabilities to process complex, fragmented, and noisy real-world datasets.
About the researcher
Dr. Peter J. Cobb is an Associate Professor in the School of Humanities and the Musketeers Foundation Institute of Data Science at the University of Hong Kong. He holds a PhD from the University of Pennsylvania and serves as the Deputy Director of the BA program in Humanities and Digital Technologies. As a field archaeologist and ceramics specialist, he directs the Ararat Plain Southeast Archaeological Project (APSAP) in Armenia, where he integrates digital humanities and data science—including augmented and mixed reality—to improve archaeological recording and the analysis of the human past.
Fund Source
Staff Seed Fund (pending)
For enquiries
please contact at atlabhku.hk
