Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

竖排文字的处理 #32

Closed
heromanba opened this issue May 15, 2020 · 5 comments
Closed

竖排文字的处理 #32

heromanba opened this issue May 15, 2020 · 5 comments

Comments

@heromanba
Copy link

您好，感谢您分享这么好的工作。想请教一下，因为CRNN这种只能处理水平的文字，PaddleOCR是如何处理竖排文字的呢？多谢！
@LDOUBLEV
Copy link
Collaborator

训练时：生成一些竖排文字样本加入训练
预测时：根据检测框的长宽比粗判断横排竖排，如果：box中 h >> w，判断为竖排，把竖排文字转成横着的去识别

@heromanba
Copy link
Author

多谢您的回复，再请问下，这个在代码里哪个地方有体现呢？我试着找了下，没找到。。

@LDOUBLEV
Copy link
Collaborator

多谢您的回复，再请问下，这个在代码里哪个地方有体现呢？我试着找了下，没找到。。

inference时： https://github.com/PaddlePaddle/PaddleOCR/blob/develop/tools/infer/predict_system.py#L55

@heromanba
Copy link
Author

十分感谢！

@adigest adigest mentioned this issue Oct 15, 2020
Closed
BillDior pushed a commit to BillDior/PaddleOCR that referenced this issue Aug 13, 2021
@hong19860320
Support syncing the libs and models during project building for andro…
0a89278 
…id demo (PaddlePaddle#32)
@Lebron-Harden
Copy link

训练时：生成一些竖排文字样本加入训练
预测时：根据检测框的长宽比粗判断横排竖排，如果：box中 h >> w，判断为竖排，把竖排文字转成横着的去识别

您好，请问训练时竖排文字样本时，图片需要预先旋转成横着的吗？

LDOUBLEV pushed a commit that referenced this issue Oct 22, 2021
@BeyondYourself
Merge pull request #32 from PaddlePaddle/release/2.3
ef1ddf8 
update 21-9-15
This was referenced Jan 11, 2023
Closed
Open
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@heromanba @LDOUBLEV @Lebron-Harden