The Benefits of Close-Domain Fine-Tuning for Table Detection in Document Images

TL;DR

This paper demonstrates that fine-tuning deep learning models for table detection using datasets of document images significantly improves accuracy compared to models fine-tuned from natural images, emphasizing the importance of domain closeness.

Contribution

The study shows that employing close-domain fine-tuning from document image datasets enhances table detection accuracy over traditional natural image pre-training.

Findings

Fine-tuning from document image datasets improves accuracy by up to 60%.

Models trained on TableBank outperform those fine-tuned from natural images.

Close-domain fine-tuning is more effective for table detection in document images.

Abstract

A correct localisation of tables in a document is instrumental for determining their structure and extracting their contents; therefore, table detection is a key step in table understanding. Nowadays, the most successful methods for table detection in document images employ deep learning algorithms; and, particularly, a technique known as fine-tuning. In this context, such a technique exports the knowledge acquired to detect objects in natural images to detect tables in document images. However, there is only a vague relation between natural and document images, and fine-tuning works better when there is a close relation between the source and target task. In this paper, we show that it is more beneficial to employ fine-tuning from a closer domain. To this aim, we train different object detection algorithms (namely, Mask R-CNN, RetinaNet, SSD and YOLO) using the TableBank dataset (a dataset of images of academic documents designed for table detection and recognition), and fine-tune them for several heterogeneous table detection datasets. Using this approach, we considerably improve the accuracy of the detection models fine-tuned from natural images (in mean a 17%, and, in the best case, up to a 60%).