Figure 1: Test image of replica Hallowe'en skull showing visualisation focused on the nasal aperture and right orbit. Note the label, 'conch'
Figure 2: Same replica Hallowe'en skull photographed on a reflective metal surface showing shifted focus to both orbits and nasal aperture. Note the label, 'mask'
Figure 3: Imaginary skulls generated by Deep Convolutional Generative Adversarial Networks, trained on 1500 images collected from Instagram
Figure 4: Imaginary skulls for sale, generated by Deep Convolutional Generative Adversarial Networks trained on 6700 images collected from Instagram
Figure 5: Screenshot of the Image Classifier App running on a smartphone. The model has been trained against images from the Portable Antiquities Scheme database, finds.org.uk. Here the classifier is 97% certain that it is seeing a fibula (displayed on the website; Graham is holding the phone in front of the monitor). The label 'brooch finds' is the same label as the folder containing the original training images, where 'finds' means it was sourced from finds.org.uk. The scale bars in the image are probably to blame for the classifier's confusion
Figure 6: Communication between collectors regarding how authenticity is determined
Internet Archaeology is an open access journal based in the Department of Archaeology, University of York. Except where otherwise noted, content from this work may be used under the terms of the Creative Commons Attribution 3.0 (CC BY) Unported licence, which permits unrestricted use, distribution, and reproduction in any medium, provided that attribution to the author(s), the title of the work, the Internet Archaeology journal and the relevant URL/DOI are given.
Internet Archaeology content is preserved for the long term with the Archaeology Data Service. Help sustain and support open access publication by donating to our Open Access Archaeology Fund.