Synthesizing Realistic Data for Table Recognition
This addresses data scarcity for table recognition in financial domains, though it appears incremental as it builds on existing synthesis approaches.
The authors tackled the problem of limited training data for table recognition by proposing a novel method to synthesize realistic annotation data using existing complex tables, creating the first extensive table annotation dataset for Chinese financial announcements and establishing the inaugural benchmark for this domain. Experiments showed comprehensive performance improvements, especially for tables with multiple spanning cells.
To overcome the limitations and challenges of current automatic table data annotation methods and random table data synthesis approaches, we propose a novel method for synthesizing annotation data specifically designed for table recognition. This method utilizes the structure and content of existing complex tables, facilitating the efficient creation of tables that closely replicate the authentic styles found in the target domain. By leveraging the actual structure and content of tables from Chinese financial announcements, we have developed the first extensive table annotation dataset in this domain. We used this dataset to train several recent deep learning-based end-to-end table recognition models. Additionally, we have established the inaugural benchmark for real-world complex tables in the Chinese financial announcement domain, using it to assess the performance of models trained on our synthetic data, thereby effectively validating our method's practicality and effectiveness. Furthermore, we applied our synthesis method to augment the FinTabNet dataset, extracted from English financial announcements, by increasing the proportion of tables with multiple spanning cells to introduce greater complexity. Our experiments show that models trained on this augmented dataset achieve comprehensive improvements in performance, especially in the recognition of tables with multiple spanning cells.