Craft-text-detector

- Relax requirements for OpenCV (25)

path for the weight file can be specified by:
python
load_craftnet_model(weight_path="path/to/weight")

python
load_refinenet_model(weight_path="path/to/weight")

* [fix empty_cuda_cache](https://github.com/fcakyon/craft-text-detector/pull/13)

* updated basic usage for better device handling, now Craft instance should be created before calling detect_text:
python
import Craft class
from craft_text_detector import Craft

set image path and export folder directory
image_path = 'figures/idcard.png'
output_dir = 'outputs/'

create a craft instance
craft = Craft(output_dir=output_dir, crop_type="poly", cuda=False)

apply craft text detection and export detected regions to output directory
prediction_result = craft.detect_text(image_path)

unload models from ram/gpu
craft.unload_craftnet_model()
craft.unload_refinenet_model()

* some internal naming and styling changes

- fix [cuda device bug](https://github.com/fcakyon/craft-text-detector/issues/8)
- fix visualization export bug

- time profiling
- better input size handling (with new long_size parameter)
- bug fixes