Writer Identification and Script Classification : Two Tasks for a Common Understanding of Cultural Heritage

International audienceWriter identification and Script classification are usually considered as two separated and very different tasks, as well in palaeography as in computer science. Following the ICDAR competition on the CLAMM corpus about script classification and dating, this contribution proposes to reconsider the tasks and methods of Palaeography and Computer Vision applied for Artificial Paleography. We argue that, while aiming at understanding past societies and the written cultural heritage in its whole complexity, palaeography and its core tasks may be defined as discretizing the historical and social continuum at different levels of granularity. In this sense, we can consider Writer identification and Script classification as one single task. We then transfer this hypothesis to computer science, by running two infrastructures created for Script classification on a more homogeneous dataset with a focus on Writer Identification. The analysis of the results confirms the unicity of the tasks and let us consider in a novel way how to evidence and illustrate the historical continuum, while discretizing it in a non-binary wa