iDocV2: Leveraging Self-Supervision and Open-Set Detection for Improving Pattern Spotting in Historical Documents

TL;DR

This paper introduces iDocV2, a self-supervised and open-set detection model that significantly improves pattern spotting accuracy and speed in historical documents, surpassing previous methods.

Contribution

The paper presents a new encoder trained with self-supervision and an open-set detector, achieving faster search and higher precision in pattern spotting tasks.

Findings

Achieves 0.612 precision on small non-square queries, surpassing previous state-of-the-art.

Reduces search time by 10 times compared to existing models.

Improves false positive reduction using non-maximum suppression.

Abstract

Considering the imminent massification of digital books, it has become critical to facilitate searching collections through graphical patterns. Current strategies for document retrieval and pattern spotting in historical documents still need to be improved. State-of-the-art strategies achieve an overall precision of $0.494$ for pattern spotting, where the precision for small non-square queries reaches 0.427. In addition, the processing time is excessive, requiring up to 7 seconds for searching in the DocExplore dataset due to a dense-based strategy used by SOTA models. Therefore, we propose a new model based on a better encoder (iDoc), trained under a self-supervised strategy, and an open-set detector to accelerate searching. Our model achieves competitive results with state-of-the-art pattern spotting and document retrieval, improving speed by 10x. Furthermore, our model reaches a new SOTA performance on the small non-square queries, achieving a new precision of 0.612.Different from the previous version, this leverages non-maximum suppression to reduce false positives.