Identifying Table Structure in Documents using Conditional Generative Adversarial Networks
This addresses the challenge of information extraction from tables in documents for industries and academic research, representing an incremental improvement over existing bottom-up methods.
The paper tackles the problem of extracting table structure from unstructured documents by proposing a top-down approach that uses a conditional generative adversarial network to map table images into a standardized skeleton form, then derives latent structure with xy-cut projection and Genetic Algorithm optimization, achieving adaptability to different configurations and requiring small training datasets.
In many industries, as well as in academic research, information is primarily transmitted in the form of unstructured documents (this article, for example). Hierarchically-related data is rendered as tables, and extracting information from tables in such documents presents a significant challenge. Many existing methods take a bottom-up approach, first integrating lines into cells, then cells into rows or columns, and finally inferring a structure from the resulting 2-D layout. But such approaches neglect the available prior information relating to table structure, namely that the table is merely an arbitrary representation of a latent logical structure. We propose a top-down approach, first using a conditional generative adversarial network to map a table image into a standardised `skeleton' table form denoting approximate row and column borders without table content, then deriving latent table structure using xy-cut projection and Genetic Algorithm optimisation. The approach is easily adaptable to different table configurations and requires small data set sizes for training.