Efficient OCR for Building a Diverse Digital History
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2023
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910542303592448 |
|---|---|
| author | Carlson, Jacob Bryan, Tom Dell, Melissa |
| author_facet | Carlson, Jacob Bryan, Tom Dell, Melissa |
| contents | Thousands of users consult digital archives daily, but the information they can access is unrepresentative of the diversity of documentary history. The sequence-to-sequence architecture typically used for optical character recognition (OCR) - which jointly learns a vision and language model - is poorly extensible to low-resource document collections, as learning a language-vision model requires extensive labeled sequences and compute. This study models OCR as a character level image retrieval problem, using a contrastively trained vision encoder. Because the model only learns characters' visual features, it is more sample efficient and extensible than existing architectures, enabling accurate OCR in settings where existing solutions fail. Crucially, the model opens new avenues for community engagement in making digital history more representative of documentary history. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2304_02737 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Efficient OCR for Building a Diverse Digital History Carlson, Jacob Bryan, Tom Dell, Melissa Computer Vision and Pattern Recognition Digital Libraries General Economics Economics Thousands of users consult digital archives daily, but the information they can access is unrepresentative of the diversity of documentary history. The sequence-to-sequence architecture typically used for optical character recognition (OCR) - which jointly learns a vision and language model - is poorly extensible to low-resource document collections, as learning a language-vision model requires extensive labeled sequences and compute. This study models OCR as a character level image retrieval problem, using a contrastively trained vision encoder. Because the model only learns characters' visual features, it is more sample efficient and extensible than existing architectures, enabling accurate OCR in settings where existing solutions fail. Crucially, the model opens new avenues for community engagement in making digital history more representative of documentary history. |
| title | Efficient OCR for Building a Diverse Digital History |
| topic | Computer Vision and Pattern Recognition Digital Libraries General Economics Economics |
| url | https://arxiv.org/abs/2304.02737 |