我的场景是需要DET提取文本行键值对
#14186
Replies: 2 comments 3 replies
-
|
无法理解你的问题 |
Beta Was this translation helpful? Give feedback.
3 replies
-
|
就是一个文本行需要检测出两个文本区域，分为键合值。比如下面这种数据 |
Beta Was this translation helpful? Give feedback.
-
|
后处理的时候根据：分割一下？ |
Beta Was this translation helpful? Give feedback.
-
|
试试PPChatOCR能不能解决你的这个需求，用大模型提取键值对 |
Beta Was this translation helpful? Give feedback.
-
|
后处理效果还行，能达到要求 |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
比如 “目标函数：DBloss” 中获取键值对“目标函数”和“DBloss” ，使用det是连续检测一个文本行的，我需要分为两个文本区域。便于进行后续的ser和re训练。目前训练效果不太好，我看pp ocrv3使用了DB检测头，使用了Vatti clipping algorithm算法，我该怎么控制文本膨胀呢？
论文中的shrunk polygon有用吗？
Beta Was this translation helpful? Give feedback.
All reactions