KaiRacters: Character-level-based Writer Retrieval for Greek Papyri

TL;DR

This paper introduces a character-level approach for Greek papyri writer retrieval, leveraging character annotations and NetVLAD to improve performance over patch-based methods, with publicly available annotated datasets.

Contribution

It presents a novel character-level feature aggregation method for writer retrieval in Greek papyri, outperforming existing patch-based approaches and providing a new annotated dataset.

Findings

Boosted performance by up to 4% mAP using character-level features

Demonstrated 11% relative improvement over baseline methods

Provided a new dataset with character-level annotations for Greek papyri

Abstract

This paper presents a character-based approach for enhancing writer retrieval performance in the context of Greek papyri. Our contribution lies in introducing character-level annotations for frequently used characters, in our case the trigram kai and four additional letters (epsilon, kappa, mu, omega), in Greek texts. We use a state-of-the-art writer retrieval approach based on NetVLAD and compare a character-level-based feature aggregation method against the current default baseline of using small patches located at SIFT keypoint locations for building the page descriptors. We demonstrate that by using only about 15 characters per page, we are able to boost the performance up to 4% mAP (a relative improvement of 11%) on the GRK-120 dataset. Additionally, our qualitative analysis offers insights into the similarity scores of SIFT patches and specific characters. We publish the dataset with character-level annotations, including a quality label and our binarized images for further research.