Acknowledgement: The cuneiform tablet fragments used in this interaction are held on study loan at the British Museum.
The authors wish to thank the Trustees of the British Museum for permission to access the fragments and Dr Jonathan Taylor whose assistance made this work possible.
About this interaction: This interaction builds on an
original version presented at the interactions gallery at the 2017 BCS HCI conference.
The interactive 3D model viewer is implemented using
WebGL with the
THREE.js JavaScript library.
The 3D models were created using structure-from-motion photogrammetric reconstruction.
Sets of 36 photographs taken using a synchronised turntable were combined with virtual calibration photographs using
VisualSFM to estimate the camera parameters.
A customised bundle adjustment algorithm refined the estimates using a-priori knowledge about the exact known properties of the virtual photographs.
Finally,
MVE was used to generate dense point clouds which were aligned and merged before the mesh was formed using
MeshLab and photographic texturing added.