How 3D models (photogrammetry) of rock art can improve recording veracity: a case study from Kakadu National Park, Australia

Abstract

Creating an inventory of a rock art site in the field can be time-consuming and expensive, but Structure-from-Motion (SfM) photogrammetry has the potential to alleviate these issues. Using SfM, rock art sites can be recorded rapidly, with a 3D model created to allow a digital inventory to be compiled. However, the veracity of a digital inventory can be questioned. At the Blue Paintings site in Kakadu National Park, Australia, we tested two field inventories against a digitally-derived inventory and ground-truthed the results. The results demonstrated that the digitally-derived inventory was slightly less comprehensive than the field recordings, but only unidentified lines and blotches were lacking; this would not necessarily adversely influence interpretation. Furthermore, the field inventories conducted by different people also had variations, demonstrating that whether the inventory is done on a 3D model or in the field, an inventory is still a human interpretation.

Keywords:

• Rock art
• photogrammetry
• Structure-from-Motion
• Kakadu
• 3D
• DStretch

Acknowledgments

This research was undertaken as part of the Pathways: people, landscape and rock art project and was funded by the Australian Research Council grant ‘Australian rock art: history, conservation and Indigenous well-being’, part of Paul Taçon’s ARC Laureate Project (FL160100123). As partners in this research, Jeffrey Lee and Parks Australia have provided much support, and we thank them for their collaboration. We are grateful to all the rangers who joined the fieldwork in 2018 and/or 2019, including Kadeem May, Gabrielle O’Loughlin, Jennifer Wellings, Jasmine Nabobbob, Duane Councillor, Joe Markham, Jacqueline Cahill, Cindy Cooper and Louise Harrison. We would also like to thank John Hayward, Janet Davill, Phil Davill, Samuel Dix, Cindy Cooper, Kadeem May, Irina Ponomareva, and Emily Miller for helping to document the Blue Paintings site and allowing us to analyse their work in this paper. We are particularly grateful to Melissa Marshall for allowing us to refer to her unpublished PhD research and providing us with further information on her documentation of Blue Paintings. We would also like to thank Sama Low-Choy for statistical advice. Finally, thanks to Paul Taçon, PERAHU and the Griffith Centre for Social and Cultural Research for supporting this research.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was funded by the Australian Research Council through the Discovery project [DP160101832] associated with the ARC Laureate Fellowship granted to Prof. Paul Taçon [FL160100123].