CRAFT-pytorch
CRAFT-pytorch copied to clipboard
clovaai
Official implementation of Character Region Awareness for Text Detection (CRAFT)
The main goal for this PR is to add a C++ binding to greatly improve the runtime of the post-processing step. This gets pretty slow as the number of words...
I am running using anaconda prompt: python test.py --trained_model=[weightfile] --test_folder='C:\Users\nosta\OneDrive\Desktop\construction_data_entry\craft_unet_tesseract_g702\test_folder' I am getting following error. import craft_utils ModuleNotFoundError: No module named 'craft_utils' Could you please help? Below is a list...
The ONNX models that I generated cannot pass in trtexec. **torch 1.4.0:** I exported with opset_version=11, otherwise it gives warning. in trtexec: ~~~ ERROR: /home/azad/TensorRT/parsers/onnx/builtin_op_importers.cpp:689 In function importGather: [8] Assertion...
hi, i tried to get max performance and fps but at the end i found that pytorch on small fixed size is better than tensorrt time for example i had...
i am a bigginer, i want kwoun the reason use vgg , or resnet is ok also.
Hi, I want to get bounding box in characters level. How can I do it ? Thanks.
both random model and refiner model used. I'm curious to refiner model.
Can you provide help/examples on how to perform training of CRAFT model?
I am not able to use Bitmap images of any size with test.py. Can you provide any advice on how to use Bitmap format images in CRAFT code?
Hi, I need to export the detected boxes from the Colab notebook, but I don't understand how to do it. I try with the code suggested by @akarazniewicz in #39,...
