What algorithm/paper does the new PaddleOCR use for detection? #15715
Answered
by
Sunting78
test3211234
asked this question in
Q&A
-
|
I know for paddleocr with paddle v2 the default algorithm used for detection was DB, but how about for newest paddle v3 paddleocr? can we choose from many algorithm options? if so what are they? |
Beta Was this translation helpful? Give feedback.
Answered by
Sunting78
Jun 13, 2025
Replies: 1 comment
-
|
In PaddleOCR v3, we still use the DB method. For the server-side model, the backbone has been replaced, while the network structure of the mobile model remains unchanged. |
Beta Was this translation helpful? Give feedback.
0 replies
Answer selected by
test3211234
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
In PaddleOCR v3, we still use the DB method. For the server-side model, the backbone has been replaced, while the network structure of the mobile model remains unchanged.