cloudfool

Results 10 issues of cloudfool

I downloaded the synthetic chinese string data from baiduyun, but can not open it. I want to know is this file corrupted or not. please give me some advice.

Hi, I want to use this model to detect Chinese text images . When I create tfrecords, the groundtruth_text for Chinese characters after 'utf-8' encoding is different from English and...

你好, 我目前还没有条件运行你的程序。 我想先问一下,这个模型可以识别长一点的文本行图片么? 我看了demo程序,里面有设置最大字符个数15,这个值是固定的么? 谢谢。

### 📚 The doc issue Hi, The first step of chatgpt is to finetune a SFT. I would like to know how to do this? Does the code already implemented...

documentation

Hi, I tested your dewarp.py and found that the dewarped image is cropped, like below: ![ZY_1](https://user-images.githubusercontent.com/22359737/131245287-c27b4cdb-5071-46e4-bbb7-af91e68febdf.jpg) ![ZY_1_0_rebook](https://user-images.githubusercontent.com/22359737/131245311-693946c9-d8b1-41f2-be94-4140907efbb6.png) I hope the result should be like this: ![ZY_1](https://user-images.githubusercontent.com/22359737/131245329-b90ffdd1-8984-43a0-858e-d402aec44293.jpg) I want to keep...

It seems that the program always loads groundtruth roi from raw gt images. If users try to change the image scale parameters in config file, it will not work.

我用tools/infer/predict_rec.py调用ch_PP-OCRv4_rec_infer模型识别下面两张图时发现识别结果分别是“冏下點”和“冏干文”,请问这是什么原因?谢谢 ![img_crop_冏下點](https://github.com/PaddlePaddle/PaddleOCR/assets/22359737/e48f8918-a856-4ec6-9aa2-279152ca7156) ![img_crop_冏干文](https://github.com/PaddlePaddle/PaddleOCR/assets/22359737/ea83245d-0356-4558-95a0-dcdba024ca31)

Hi, I got this following error when I run the code: -- Process 0 terminated with the following error: Traceback (most recent call last): File "/opt/conda/lib/python3.7/site-packages/torch/multiprocessing/spawn.py", line 20, in _wrap...

你好,请问xml格式的发票可以转pdf吗?

hi, 我这边有一份ofd的文件,转成图片之后 很奇怪 ![image](https://github.com/user-attachments/assets/33a27e2a-1c32-46f3-a1db-1c57f11d6d77) 原ofd件已经发您的邮箱了,请看一下。 这份文件的特别之处是感觉xml里面存储的都是笔画而不是文字和文字框,能给我提供一些建议吗?谢谢