mboillet commited on
Commit
96011bc
1 Parent(s): 95cd091

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +14 -10
README.md CHANGED
@@ -24,12 +24,12 @@ This model performs Handwritten Text Recognition in Chinese.
24
 
25
  ## Model description
26
 
27
- The model was trained using the PyLaia library on the [CASIA-HWDB2](http://www.nlpr.ia.ac.cn/databases/handwriting/Offline_database.html) document images.
28
 
29
  Training images were resized with a fixed height of 128 pixels, keeping the original aspect ratio.
30
 
31
- | split | N lines |
32
- | ----- | ------: |
33
  | train | 33,425 |
34
  | val | 8,325 |
35
  | test | 10,449 |
@@ -40,22 +40,26 @@ An external 6-gram character language model can be used to improve recognition.
40
 
41
  The model achieves the following results:
42
 
43
- | set | Language model | CER (%) | N lines |
44
  |:------|:---------------| ----------:|----------:|
45
  | test | no | 4.61 | 10,449 |
46
  | test | yes | 1.53 | 10,449 |
47
 
48
  ## How to use?
49
 
50
- Please refer to the [documentation](https://atr.pages.teklia.com/pylaia/).
51
 
52
  ## Cite us!
53
 
54
  ```bibtex
55
- @inproceedings{pylaia-lib,
56
- author = "Tarride, Solène and Schneider, Yoann and Generali, Marie and Boillet, Melodie and Abadie, Bastien and Kermorvant, Christopher",
57
- title = "Improving Automatic Text Recognition with Language Models in the PyLaia Open-Source Library",
58
- booktitle = "Submitted at ICDAR2024",
59
- year = "2024"
 
 
 
 
60
  }
61
  ```
 
24
 
25
  ## Model description
26
 
27
+ The model was trained using the PyLaia library on the [CASIA-HWDB2](http://www.nlpr.ia.ac.cn/databases/handwriting/Offline_database.html) dataset.
28
 
29
  Training images were resized with a fixed height of 128 pixels, keeping the original aspect ratio.
30
 
31
+ | set | lines |
32
+ | :----- | ------: |
33
  | train | 33,425 |
34
  | val | 8,325 |
35
  | test | 10,449 |
 
40
 
41
  The model achieves the following results:
42
 
43
+ | set | Language model | CER (%) | lines |
44
  |:------|:---------------| ----------:|----------:|
45
  | test | no | 4.61 | 10,449 |
46
  | test | yes | 1.53 | 10,449 |
47
 
48
  ## How to use?
49
 
50
+ Please refer to the [PyLaia documentation](https://atr.pages.teklia.com/pylaia/usage/prediction/) to use this model.
51
 
52
  ## Cite us!
53
 
54
  ```bibtex
55
+ @inproceedings{pylaia2024,
56
+ author = {Tarride, Solène and Schneider, Yoann and Generali-Lince, Marie and Boillet, Mélodie and Abadie, Bastien and Kermorvant, Christopher},
57
+ title = {{Improving Automatic Text Recognition with Language Models in the PyLaia Open-Source Library}},
58
+ booktitle = {Document Analysis and Recognition - ICDAR 2024},
59
+ year = {2024},
60
+ publisher = {Springer Nature Switzerland},
61
+ address = {Cham},
62
+ pages = {387--404},
63
+ isbn = {978-3-031-70549-6}
64
  }
65
  ```