TextileNet: Towards Zero-shot Text-style Segmentation of Manuscripts
For paleographers and digital humanities researchers, this work addresses the lack of labeled training data and open scribe sets by enabling zero-shot transfer to manuscript analysis, though results are preliminary and the gender identification finding is cautionary.
TextileNet uses a fully convolutional multi-task network trained on synthetic data to produce pixel-level texture embeddings for zero-shot script-style segmentation of historical manuscripts, achieving credible results in hand and gender identification while cautioning that gender attribution in handwriting requires careful interpretation.
Automatic writer identification systems have progressed remarkably in recent years, yet their deployment in archival paleography remains limited by the scarcity of labeled training data, open scribe sets, and degraded image quality. We present TextileNet, a fully convolutional multi-task network trained exclusively on synthetic data to produce dense pixel-level texture embeddings, which we transfer zeroshot to historical manuscript analysis. As an original contribution to evaluation methodology, we designed a paleographic visual quiz of 80 pair and triplet questions and administered it to a range from lay participants to senior paleographers under strict anonymity, establishing to our knowledge for the first time a human baseline for script-style discrimination on late medieval text. We employ TextileNet embeddings to perform zero-shot retrieval on sub-word granularity for hand and gender identification. Our experimental results help in building the credibility of TextileNet in the paleographic domain, but more than that demonstrate in experimental terms that the question of gender in handwriting needs to be treated with caution.