Skip to content

Inquiry regarding pre-trained weights for Korean (Hangul) #3

Description

@minncha

Hi, thank you for sharing this impressive work on VitSTR.

My name is Minchan Ahn, a researcher at the Computer Vision Lab at Inha University. I am highly interested in applying your model to Korean text recognition tasks.

I was wondering if you (or anyone in the community) have experimented with Korean characters (Hangul) or have any pre-trained checkpoints available for Korean datasets?

If public weights are not available, could you share any specific configurations or tips that might be helpful for training VitSTR on a language with a large character set like Korean?

Thanks in advance for your help.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions