TableFormer: Table Structure Understanding with Transformers
This work addresses the challenge of extracting structured data from diverse table formats for systems like search engines and knowledge graphs, representing an incremental improvement over existing methods.
The paper tackles the problem of table structure identification from images by introducing TableFormer, which improves the state-of-the-art model by using a new object detection decoder for table-cells and transformer-based decoders, resulting in TEDS scores increasing from 91% to 98.5% on simple tables and from 88.7% to 95% on complex tables.
Tables organize valuable content in a concise and compact representation. This content is extremely valuable for systems such as search engines, Knowledge Graph's, etc, since they enhance their predictive capabilities. Unfortunately, tables come in a large variety of shapes and sizes. Furthermore, they can have complex column/row-header configurations, multiline rows, different variety of separation lines, missing entries, etc. As such, the correct identification of the table-structure from an image is a non-trivial task. In this paper, we present a new table-structure identification model. The latter improves the latest end-to-end deep learning model (i.e. encoder-dual-decoder from PubTabNet) in two significant ways. First, we introduce a new object detection decoder for table-cells. In this way, we can obtain the content of the table-cells from programmatic PDF's directly from the PDF source and avoid the training of the custom OCR decoders. This architectural change leads to more accurate table-content extraction and allows us to tackle non-english tables. Second, we replace the LSTM decoders with transformer based decoders. This upgrade improves significantly the previous state-of-the-art tree-editing-distance-score (TEDS) from 91% to 98.5% on simple tables and from 88.7% to 95% on complex tables.