Teklia/doc-ufcn-norhand-v1-line

Doc-UFCN - NorHand v1 - Line detection

The NorHand v1 line detection model predicts the following elements from NorHand document images:

• vertical text lines;
• horizontal text lines.

This model was developed during the HUGIN-MUNIN project.

Model description

The model has been trained using the Doc-UFCN library on the NorHand dataset. It has been trained on images with their largest dimension equal to 768 pixels, keeping the original aspect ratio.

Evaluation results

The model achieves the following results:

set	class	IoU	F1	AP@[.5]	AP@[.75]	AP@[.5,.95]
train	vertical	88.29	89.67	71.37	33.26	36.32
	horizontal	69.81	81.35	91.73	36.62	45.67
val	vertical	73.01	75.13	46.02	4.99	15.58
	horizontal	61.65	75.69	87.98	11.18	31.55
test	vertical	78.62	80.03	59.93	15.90	24.11
	horizontal	63.59	76.49	95.93	24.18	41.45

How to use?

Please refer to the Doc-UFCN library page to use this model.

Cite us!

@inproceedings{doc_ufcn2021,
    author = {Boillet, Mélodie and Kermorvant, Christopher and Paquet, Thierry},
    title = {{Multiple Document Datasets Pre-training Improves Text Line Detection With
              Deep Neural Networks}},
    booktitle = {2020 25th International Conference on Pattern Recognition (ICPR)},
    year = {2021},
    month = Jan,
    pages = {2134-2141},
    doi = {10.1109/ICPR48806.2021.9412447}
}