Lope/NoLope Dataset

The corpus comprises line images (and labels at character-level) extracted from 39 manuscripts handwritten by Lope de Vega and 223 manuscripts in various other scripts. A precise count of the distinct non-Lope hands is challenging to determine without a dedicated paleographic analysis for each document; however, it is estimated that these materials represent approximately 100 to 300 different non-Lope hands. The manuscripts were sourced from multiple institutions—including the Spanish National Library, the British Library and the Institut del Teatre of Barcelona, and were digitized through a range of processes, such as high-resolution professional reproductions, low-resolution black-and-white imaging, and photographs taken by the research team. For more details about this dataset, refer to: Toselli, A.H., Cuéllar, Á., Boadas, S., Vidal, E., Sánchez, J.A. Writer Identification Using Simplified Handwritten Text Recognition Models. In: Gonçalves, N., Oliveira, H.P., Sánchez, J.A. (eds) Pattern Recognition and Image Analysis. IbPRIA 2025. Lecture Notes in Computer Science, vol 15937. Springer, Cham. https://doi.org/10.1007/978-3-031-99565-1_20