Datasets:
Tasks:
Image-to-Text
Formats:
imagefolder
Sub-tasks:
image-captioning
Languages:
English
Size:
10K - 100K
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
This dataset we prepared using the Scanned receipts OCR and information extraction(SROIE) dataset.
|
| 2 |
The SROIE dataset contains 973 scanned receipts in English language.
|
| 3 |
Cropping the bounding boxes from each of the receipts to generate this text-recognition dataset resulted in 33626 images for train set and 18704 images for the test set.
|
|
|
|
| 1 |
+
---
|
| 2 |
+
annotations_creators: []
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
language_creators: []
|
| 6 |
+
license: []
|
| 7 |
+
multilinguality:
|
| 8 |
+
- monolingual
|
| 9 |
+
pretty_name: SROIE_2019_text_recognition
|
| 10 |
+
size_categories: []
|
| 11 |
+
source_datasets: []
|
| 12 |
+
tags:
|
| 13 |
+
- text-recognition
|
| 14 |
+
- recognition
|
| 15 |
+
task_categories:
|
| 16 |
+
- image-to-text
|
| 17 |
+
task_ids:
|
| 18 |
+
- image-captioning
|
| 19 |
+
---
|
| 20 |
This dataset we prepared using the Scanned receipts OCR and information extraction(SROIE) dataset.
|
| 21 |
The SROIE dataset contains 973 scanned receipts in English language.
|
| 22 |
Cropping the bounding boxes from each of the receipts to generate this text-recognition dataset resulted in 33626 images for train set and 18704 images for the test set.
|