How to recognize Chinese characters

[Betty-J]

Prerequisite

• I have searched Issues and Discussions but cannot get the expected help.
• The bug has not been fixed in the latest version (0.x) or latest version (1.x).

Task

I'm using the official example scripts/configs for the officially supported tasks/models/datasets.

Branch

main branch https://github.com/open-mmlab/mmocr

Environment

Follow the official documentation

Reproduces the problem - code sample

Follow the official documentation

Reproduces the problem - command or script

Follow the official documentation，https://mmocr.readthedocs.io/en/dev-1.x/user_guides/inference.html

Reproduces the problem - error message

I downloaded this model and tried to load it, but got an error：https://github.com/open-mmlab/mmocr/blob/0.x/configs/textrecog/sar/README.md#chinese-dataset


As a result, no characters are detected. How to solve this problem?

Additional information

No response