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

arXiv: https://arxiv.org/abs/2407.07536 · HTML: https://mlsanity.com/paper/2407.07536/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 20 references that are in the map.
- First posted on arXiv 10 Jul 2024.

## Builds on (most cited first, 7 of 7)

- [Unsupervised Feature Learning for Writer Identification and Writer Retrieval](https://mlsanity.com/paper/1705.09369/) (2017)
- [Encoding CNN Activations for Writer Recognition](https://mlsanity.com/paper/1712.07923/) (2017)
- [Generalized Max Pooling](https://mlsanity.com/paper/1406.0312/) (2014)
- [Writer Identification and Writer Retrieval Based on NetVLAD with Re-ranking](https://mlsanity.com/paper/2012.06186/) (2020)
- [Writer Retrieval and Writer Identification in Greek Papyri](https://mlsanity.com/paper/2212.07664/) (2022)
- [Towards Writer Retrieval for Historical Datasets](https://mlsanity.com/paper/2305.05358/) (2023)
- [Feature Mixing for Writer Retrieval and Identification on Papyri Fragments](https://mlsanity.com/paper/2306.12939/) (2023)

## Built on by (most cited first, 2 of 2)

- [Towards the Influence of Text Quantity on Writer Retrieval](https://mlsanity.com/paper/2506.07566/) (2025)
- [Assessing the impact of Binarization for Writer Identification in Greek Papyrus](https://mlsanity.com/paper/2506.15852/) (2025)

## Abstract (arXiv)

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.