Update README.md
Browse files
README.md
CHANGED
@@ -16,7 +16,6 @@ metrics:
|
|
16 |
pipeline_tag: image-segmentation
|
17 |
---
|
18 |
|
19 |
-
|
20 |
# Doc-UFCN - Generic page detection
|
21 |
|
22 |
The generic page detection model predicts single pages from document images.
|
@@ -30,20 +29,20 @@ It has been trained on images with their largest dimension equal to 768 pixels,
|
|
30 |
|
31 |
The model achieves the following results:
|
32 |
|
33 |
-
|
|
34 |
-
|
|
35 |
| HOME | test | 93.92 | 95.84 | 98.98 | 98.98 | 97.61 |
|
36 |
| Horae | test | 96.68 | 98.31 | 99.76 | 98.49 | 98.08 |
|
37 |
| Horae | test-300 | 95.66 | 97.27 | 98.87 | 98.45 | 97.38 |
|
38 |
|
39 |
## How to use?
|
40 |
|
41 |
-
Please refer to the Doc-UFCN library page
|
42 |
|
43 |
## Cite us!
|
44 |
|
45 |
```bibtex
|
46 |
-
@inproceedings{
|
47 |
author = {Boillet, Mélodie and Kermorvant, Christopher and Paquet, Thierry},
|
48 |
title = {{Multiple Document Datasets Pre-training Improves Text Line Detection With
|
49 |
Deep Neural Networks}},
|
|
|
16 |
pipeline_tag: image-segmentation
|
17 |
---
|
18 |
|
|
|
19 |
# Doc-UFCN - Generic page detection
|
20 |
|
21 |
The generic page detection model predicts single pages from document images.
|
|
|
29 |
|
30 |
The model achieves the following results:
|
31 |
|
32 |
+
| dataset | set | IoU | F1 | AP@[.5] | AP@[.75] | AP@[.5,.95] |
|
33 |
+
| :----- | :------- | ----: | ----: | ------: | -------: | ----------: |
|
34 |
| HOME | test | 93.92 | 95.84 | 98.98 | 98.98 | 97.61 |
|
35 |
| Horae | test | 96.68 | 98.31 | 99.76 | 98.49 | 98.08 |
|
36 |
| Horae | test-300 | 95.66 | 97.27 | 98.87 | 98.45 | 97.38 |
|
37 |
|
38 |
## How to use?
|
39 |
|
40 |
+
Please refer to the [Doc-UFCN library page](https://pypi.org/project/doc-ufcn/) to use this model.
|
41 |
|
42 |
## Cite us!
|
43 |
|
44 |
```bibtex
|
45 |
+
@inproceedings{doc_ufcn2021,
|
46 |
author = {Boillet, Mélodie and Kermorvant, Christopher and Paquet, Thierry},
|
47 |
title = {{Multiple Document Datasets Pre-training Improves Text Line Detection With
|
48 |
Deep Neural Networks}},
|