Identifying Cross-Depicted Historical Motifs

TL;DR

This paper demonstrates that deep learning techniques can effectively identify the same object across different depiction styles in historical documents, outperforming traditional methods.

Contribution

It introduces a deep learning approach for cross-depiction recognition in historical watermarks, achieving high accuracy and robustness on classification and similarity tasks.

Findings

Deep CNNs achieved 96% classification accuracy.

False positive rate at 95% TPR is 0.11.

Outperforms state-of-the-art methods significantly.

Abstract

Cross-depiction is the problem of identifying the same object even when it is depicted in a variety of manners. This is a common problem in handwritten historical documents image analysis, for instance when the same letter or motif is depicted in several different ways. It is a simple task for humans yet conventional heuristic computer vision methods struggle to cope with it. In this paper we address this problem using state-of-the-art deep learning techniques on a dataset of historical watermarks containing images created with different methods of reproduction, such as hand tracing, rubbing, and radiography. To study the robustness of deep learning based approaches to the cross-depiction problem, we measure their performance on two different tasks: classification and similarity rankings. For the former we achieve a classification accuracy of 96% using deep convolutional neural networks. For the latter we have a false positive rate at 95% true positive rate of 0.11. These results outperform state-of-the-art methods by a significant margin.