Artificial Intelligence, 3D Documentation, and Rock Art—Approaching and Reflecting on the Automation of Identification and Classification of Rock Art Images

Abstract Rock art carvings, which are best described as petroglyphs, were produced by removing parts of the rock surface to create a negative relief. This tradition was particularly strong during Nordic Bronze Age (1700–550 BC) in southern Scandinavia with over 20,000 boats and thousands humans, animals, wagons, etc . vivid highly engaging material provides quantitative data high potential understand social structures ideologies. The ability provide technically possible documentation automate identification classification images would help take full advantage research petroglyphs elsewhere. We, therefore, attempted train model that locates classifies image objects using faster region-based convolutional neural network (Faster-RCNN) based on novel method improve visualizing content 3D documentations. A newly created layer currently available has reduced inscribed bias compared older methods. Several models trained input annotated bounding boxes different parameters find solution. included 4305 individual 408 scans sites. To enhance enrich training data, we used augmentation transfer learning. successful perform exceptionally well circles, human figures wheels. work an interdisciplinary undertaking led important reflections about archaeology, digital humanities, artificial intelligence. success represented open avenues for future art.