Stanislav Kalinin
commited on
Commit
•
76966f3
1
Parent(s):
ddc318b
docs: Add readme
Browse files- README.md +54 -0
- pics/counts_hor.png +3 -0
- pics/leaderboard.jpeg +3 -0
- pics/line_label.jpeg +3 -0
- pics/one_line.jpg +3 -0
README.md
ADDED
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# Digital Peter
|
2 |
+
|
3 |
+
Paper Is available at http://arxiv.org/abs/2103.09354
|
4 |
+
|
5 |
+
## Description
|
6 |
+
|
7 |
+
Digital Peter is an educational task with a historical slant created on the basis of several AI technologies (Computer Vision, NLP, and knowledge graphs). The task was prepared jointly with the Saint Petersburg Institute of History (N.P.Lihachov mansion) of Russian Academy of Sciences, Federal Archival Agency of Russia and Russian State Archive of Ancient Acts.
|
8 |
+
|
9 |
+
A detailed description of the problem (with an immersion in the problem) can be found in [detailed_description_of_the_task_en.pdf`](https://github.com/sberbank-ai/digital_peter_aij2020/blob/master/desc/detailed_description_of_the_task_en.pdf)
|
10 |
+
|
11 |
+
The dataset consists of 662 full page images and 9696 annotated text files. There are 265788 symbols and approximately 50998 words.
|
12 |
+
|
13 |
+
<p align="center">
|
14 |
+
<img src="pics/counts_hor.png" width="80%">
|
15 |
+
</p>
|
16 |
+
|
17 |
+
|
18 |
+
Here is an example of one line of text.
|
19 |
+
|
20 |
+
<p align="center">
|
21 |
+
<img src="pics/one_line.jpg" width="70%">
|
22 |
+
</p>
|
23 |
+
|
24 |
+
Here is an example of segmented document.
|
25 |
+
|
26 |
+
<p align="center">
|
27 |
+
<img src="pics/line_label.jpeg" width="60%">
|
28 |
+
</p>
|
29 |
+
|
30 |
+
|
31 |
+
## Annotation format
|
32 |
+
|
33 |
+
The annotation is in COCO format. The `annotation.json` should have the following dictionaries:
|
34 |
+
|
35 |
+
- `annotation["categories"]` - a list of dicts with a categories info (categotiy names and indexes).
|
36 |
+
- `annotation["images"]` - a list of dictionaries with a description of images, each dictionary must contain fields:
|
37 |
+
- `file_name` - name of the image file.
|
38 |
+
- `id` for image id.
|
39 |
+
- `annotation["annotations"]` - a list of dictioraties with a murkup information. Each dictionary stores a description for one polygon from the dataset, and must contain the following fields:
|
40 |
+
- `image_id` - the index of the image on which the polygon is located.
|
41 |
+
- `category_id` - the polygon’s category index.
|
42 |
+
- ```attributes``` - dict with some additional annotatioin information. In the `translation` subdict you can find text translation for the line.
|
43 |
+
- `segmentation` - the coordinates of the polygon, a list of numbers - which are coordinate pairs x and y.
|
44 |
+
|
45 |
+
## Competition
|
46 |
+
|
47 |
+
We held a competition based on Digital Peter dataset.
|
48 |
+
Here is github [link](https://github.com/sberbank-ai/digital_peter_aij2020). Here is competition [page](https://ods.ai/tracks/aij2020) (need to register).
|
49 |
+
|
50 |
+
Here is the final leaderboard for this competition. Scores are presented for the private set.
|
51 |
+
|
52 |
+
<p align="center">
|
53 |
+
<img src="pics/leaderboard.jpeg" width="70%">
|
54 |
+
</p>
|
pics/counts_hor.png
ADDED
Git LFS Details
|
pics/leaderboard.jpeg
ADDED
Git LFS Details
|
pics/line_label.jpeg
ADDED
Git LFS Details
|
pics/one_line.jpg
ADDED
Git LFS Details
|