PaddleOCR icon indicating copy to clipboard operation
PaddleOCR copied to clipboard

paddle text detection set custom model

Open R4Ajeti opened this issue 3 years ago • 0 comments

请提供下述完整信息以便快速定位问题/Please provide the following information to quickly locate the problem Hi all, I am trying to add models from here: https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.6/doc/doc_en/models_list_en.md#1.2

Code currently i'm using: `from paddleocr import PaddleOCR, draw_ocr import cv2 ocr = PaddleOCR(use_angle_cls=True, lang='en', show_log = False) frame=cv2.imread("test.jpg")

result = ocr.ocr(frame, det=True, rec=False, cls=True)`

Is there and settings or simple code how to add this english model name "en_PP-OCRv3_det_slim" (i wana use slim model so detection can run faster on my images) I need only text detection. Or can i use paddle slim to do text detection faster? If yes how to get started in python?

Pls help i am new here, Happy coding :)

  • 系统环境/System Environment:ubuntu 20.04
  • 版本号/Version:Paddle: PaddleOCR: 问题相关组件/Related components:paddle==2.4.0rc0, paddleocr==2.6.1.0
  • 运行指令/Command Code:python script
  • 完整报错/Complete Error Message:

R4Ajeti avatar Nov 09 '22 19:11 R4Ajeti