Form Understanding
Form understanding research focuses on automatically interpreting the structure and content of forms, whether handwritten, scanned, or digital, to enable efficient data extraction and processing. Current efforts leverage deep neural networks, including transformers and graph neural networks, to model both the visual layout and semantic content of forms, often incorporating hierarchical structures and multi-granularity approaches for improved accuracy. This field is crucial for automating document processing across various sectors, improving accessibility for diverse languages, and facilitating the design of more intuitive and efficient forms themselves.
Papers
July 6, 2024
June 15, 2024
June 13, 2024
March 6, 2024
February 7, 2024
May 8, 2023
March 9, 2023
January 11, 2023
November 10, 2022