---
license: mit
language:
- en
- ru
task_categories:
- image-to-text
tags:
- text-scene-detection
- text-detection
- yolo
size_categories:
- 10K<n<100K
---
A text scene detection dataset, manually labeled in yolo format and contains 25 197 images, mostly with Latin and Russian text.
<p align="left">

<img src="https://cdn-uploads.huggingface.co/production/uploads/63802348f6d1901fcab06bd5/8M0XczbE7L7dDU6l-uySn.jpeg"  width=1024 height=1024/>
 
</p>