Image Based 3D Posture Matching in Real Time for Stone Tool Assembly

Through the study of stone tools in Japan, archaeologists can gain insights into the behaviors of Paleolithic people. By joining materials, stone tools can be reassembled into their original form. This assembly is a crucial and unavoidable process in archeology because it is essential for analyzing the method of how to make a stone tool. Through trial and error, archaeologists attempt to recreate and map the structures of joining materials. When pieces of joining material are investigated after it is disassembled, archaeologists must reassemble them. Therefore, the identification of stone tools and visualization of their joining order is crucial for assembling easily. In this paper, we propose a faster and more efficient approach to identifying target stone tools using stable posture estimation and matching. The proposed method is applied to real objects and offers a novel approach to identification and reassembly in archeology.