如何将识别结果导出为markdown格式？ - RapidOCR 文档 #640
Unanswered
giscus[bot]
Bot
Replies: 2 comments 3 replies
-
|
好奇这个，是用坐标信息重新排版吗 |
Beta Was this translation helpful? Give feedback.
2 replies
-
|
是的。可以参见这部分源码：link |
Beta Was this translation helpful? Give feedback.
-
|
奥，感谢，最近在做发票识别，也想用坐标还原票面 |
Beta Was this translation helpful? Give feedback.
-
|
支持类似于minuerU标签分类吗。那样好恢复格式。纯ocr提取是难以保证原格式的 |
Beta Was this translation helpful? Give feedback.
1 reply
-
|
这个不支持哈。我想你说的应该是给出每一块是表格，还是段落，这种标签吧！这个是版面分析模型搞得。 可以看一下RapidDoc项目：https://github.com/RapidAI/RapidDoc |
Beta Was this translation helpful? Give feedback.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
如何将识别结果导出为markdown格式？ - RapidOCR 文档
Awesome OCR multiple programing languages toolkits based on ONNXRuntime, OpenVION and PaddlePaddle.
https://rapidai.github.io/RapidOCRDocs/main/install_usage/rapidocr/how_to_convert_to_markdown/
Beta Was this translation helpful? Give feedback.
All reactions