PyPottery: an AI-powered end-to-end suite for pottery processing and publication
PyPottery reports a median 40x perceived speedup for turning hand-drawn pottery records into publishable material.
The open-source suite links four modules covering scan extraction, handwriting recognition, inking, semantic vectorization, and layout generation. It was evaluated on 50 hand-drawn sheets with 240 pottery drawings from Terramara di Montale in Italy. The paper frames the system as semi-automation, with archaeologists’ effort shifted toward review and higher-level documentation work. ArXiv · AI/CL/LG's note
The open-source suite links four modules covering scan extraction, handwriting recognition, inking, semantic vectorization, and layout generation. It was evaluated on 50 hand-drawn sheets with 240 pottery drawings from Terramara di Montale in Italy. The paper frames the system as semi-automation, with archaeologists’ effort shifted toward review and higher-level documentation work. ArXiv · AI/CL/LG's note
score 4