add rec image shape note for v3 rec

This commit is contained in:
andyjpaddle
2022-04-29 06:26:03 +00:00
parent 14831901af
commit 16eb7516b4
8 changed files with 22 additions and 0 deletions
+2
View File
@@ -208,6 +208,8 @@ Execute the built executable file:
./build/ppocr [--param1] [--param2] [...]
```
**Note**:ppocr uses the `PP-OCRv3` model by default, and the input shape used by the recognition model is `3, 48, 320`, so if you use the recognition function, you need to add the parameter `--rec_img_h=48`, if you do not use the default `PP-OCRv3` model, you do not need to set this parameter.
Specifically,
##### 1. det+cls+rec:
+3
View File
@@ -213,6 +213,9 @@ CUDNN_LIB_DIR=/your_cudnn_lib_dir
本demo支持系统串联调用,也支持单个功能的调用,如,只使用检测或识别功能。
**注意** ppocr默认使用`PP-OCRv3`模型,识别模型使用的输入shape为`3,48,320`, 因此如果使用识别功能,需要添加参数`--rec_img_h=48`,如果不使用默认的`PP-OCRv3`模型,则无需设置该参数。
运行方式:
```shell
./build/ppocr [--param1] [--param2] [...]