French Census

Research on French census data focuses on developing efficient methods for extracting information from historical records, often handwritten, using techniques like end-to-end information extraction and handwritten table recognition. Current efforts leverage deep learning models, including Transformers and U-Nets, often incorporating attention mechanisms to improve accuracy and efficiency in tasks such as text recognition, information extraction, and even gender detection within audio-visual media. This work facilitates large-scale historical analysis, enabling deeper understanding of French social and economic structures over time and providing valuable resources for researchers in demography, history, and the humanities.