-
Name: paddleocr |
Beta Was this translation helpful? Give feedback.
Answered by
qianliyx
Sep 12, 2024
Replies: 3 comments 2 replies
-
使用 #10377 的方案并不能很好的解决这个问题。如场景是影印版本的pdf,这个效果定位会更加不准确。下图是我将CTC返回的col位置索引 * cellWh (cellWh = box-width / col-len)得到的每个字符的初识坐标,不管在中文、符号或数字的情况下,都并未呈现出某种一致的规律,请问CTC返回的这个所谓的位置,底层到底是什么样的逻辑~ |
Beta Was this translation helpful? Give feedback.
1 reply
-
可以试试我们在RapidOCR中集成的PaddleOCR单字坐标:https://github.com/RapidAI/RapidOCR/releases/tag/v1.4.0 |
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
设置:return_word_box = True时,返回:
[[[26.0, 37.0], [304.0, 37.0], [304.0, 73.0], [26.0, 73.0]], ('纯臻营养护发素', 0.9946897625923157, [46.085826210826205, [['纯', '臻', '营', '养', '护', '发', '素']], [[3, 10, 16, 23, 30, 36, 43]], ['cn']])]
请问这个返回怎么使用,有没有说明,怎么根据这个返回获取单字具体坐标呢?