Forums

Mobile-Captured Image Document Recognition for Vietnamese Receipts (MC-OCR) Forum

Go back to competition Back to thread list Post in this thread

> Can i use data augmentation or synthetic data for text recognition task

Can i use data augmentation or synthetic data for text recognition task

Posted by: ducnt97 @ Dec. 30, 2020, 3:03 a.m.

Hi ducnt97,

Only manual annotation is NOT allowed.
As long as the process is automated, you're good to go.

Regards,

Posted by: mc-ocr-organizers @ Dec. 31, 2020, 4:46 p.m.

Hi organizers,
No manual annotation that mean we can use pretrained models to automatically annotate data?

Posted by: SDSV_AICR @ Jan. 2, 2021, 5:31 a.m.

Hi ducnt97,
The challenge provides public data to finetune pretrained models or train models from scratch. That’s the standard process for any data challenge.

Best regards,
/The MC-OCR Team

Posted by: mc-ocr-organizers @ Jan. 3, 2021, 6:18 p.m.
Post in this thread