Hugging Face's logo

Datasets:
priyank-m
/
SROIE_2019_text_recognition

Tasks:
Image-to-Text
Modalities:
Image
Text
Formats:
imagefolder
Sub-tasks:
image-captioning
Languages:
English
Size:
10K - 100K
Tags:
text-recognition
recognition
Libraries:
Datasets
Dataset card Data Studio Files Community
Dataset Viewer
Auto-converted to Parquet
The dataset viewer is taking too long to fetch the data. Try to refresh this page.
Server-side error
Error code:   ClientConnectionError

This dataset we prepared using the Scanned receipts OCR and information extraction(SROIE) dataset. The SROIE dataset contains 973 scanned receipts in English language. 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. The text annotations for all the images inside a split are stored in a metadata.jsonl file.

usage:

from dataset import load_dataset

data = load_dataset("priyank-m/SROIE_2019_text_recognition")

source of raw SROIE dataset: https://www.kaggle.com/datasets/urbikn/sroie-datasetv2

Downloads last month
285
Size of downloaded dataset files:
183 MB
Size of the auto-converted Parquet files:
163 MB
Number of rows:
52,330

Collection including priyank-m/SROIE_2019_text_recognition