Views
No views yet
| set | lines |
|---|---|
| train | 25,800 |
| val | 3,102 |
| test | 3,819 |
| set | Language model | CER (%) | WER (%) | lines |
|---|---|---|---|---|
| test | no | 10.54 | 28.12 | 3,819 |
| test | yes | 9.52 | 23.73 | 3,819 |
1@inproceedings{pylaia2024,
2 author = {Tarride, Solène and Schneider, Yoann and Generali-Lince, Marie and Boillet, Mélodie and Abadie, Bastien and Kermorvant, Christopher},
3 title = {{Improving Automatic Text Recognition with Language Models in the PyLaia Open-Source Library}},
4 booktitle = {Document Analysis and Recognition - ICDAR 2024},
5 year = {2024},
6 publisher = {Springer Nature Switzerland},
7 address = {Cham},
8 pages = {387--404},
9 isbn = {978-3-031-70549-6}
10}1@inproceedings{belfort-2023,
2 author = {Tarride, Solène and Faine, Tristan and Boillet, Mélodie and Mouchère, Harold and Kermorvant, Christopher},
3 title = {Handwritten Text Recognition from Crowdsourced Annotations},
4 year = {2023},
5 isbn = {9798400708411},
6 publisher = {Association for Computing Machinery},
7 address = {New York, NY, USA},
8 url = {https://doi.org/10.1145/3604951.3605517},
9 doi = {10.1145/3604951.3605517},
10 booktitle = {Proceedings of the 7th International Workshop on Historical Document Imaging and Processing},
11 pages = {1–6},
12 numpages = {6},
13 keywords = {Crowdsourcing, Handwritten Text Recognition, Historical Documents, Neural Networks, Text Aggregation},
14 series = {HIP '23}
15}