CRAFT / README.md
arxyzan's picture
Update README.md
0a071ab verified
|
raw
history blame
778 Bytes
---
language:
- fa
- en
library_name: hezar
pipeline_tag: object-detection
tags:
- hezar
---
## CRAFT: Character-Region Awareness For Text detection
CRAFT is a multilingual text detection model. The original implementation is located at https://github.com/clovaai/CRAFT-pytorch. This repo is only compatible
with the [Hezar](https://github.com/hezarai/hezar) package.
### Usage
```
pip install hezar[vision]
```
```python
import cv2
from hezar.models import Model
from hezar.utils import draw_boxes
model = Model.load("hezarai/CRAFT", device="cuda")
outputs = model.predict("../assets/text_detection_example.png")
image = cv2.imread("../assets/text_detection_example.png")
result_image = draw_boxes(image, outputs[0]["boxes"])
cv2.imwrite("detected.png", result_image)
```