Transferable object detection approaches in archaeology for both terrestrial- and underwater-based projects

Summary

This is an abstract from the "Practice, Theory, and Ethics of Machine Learning in Archaeology" session, at the 90th annual meeting of the Society for American Archaeology.

This talk will focus on deep learning approaches to object detection in archaeology using remotely sensed data. We will discuss several case studies that use similar methodological approaches, presenting shared conclusions drawn from across the case studies. Case studies will include two terrestrial projects focused on ancient Maya features and ancient Romanian burial mounds using lidar and RGB imagery, respectively, as well as two underwater projects, focused on shipwrecks and aircraft wrecks using multibeam sonar and sidescan sonar, respectively. Presented methodological conclusions will apply to the deep learning approach in general, irrespective of study area or feature type, and will include discussion of deep learning modeling, data and imagery collection parameters where applicable, data and imagery preprocessing, model assessment, applicability and relevance of approach, and recommended implementation.

This Resource is Part of the Following Collections

• Society for American Archaeology 90th Annual Meeting, Denver, CO (2025)   •
• Practice, Theory, and Ethics of Machine Learning in Archaeology

Keywords

Record Identifiers

Abstract Id(s): 51938