deep-text-recognition-benchmark
deep-text-recognition-benchmark copied to clipboard
Special Characters Recognition
Hi, I am trying t use the ST_SPE.zip lmdb dataset of the special characters to fine-tune another model which has been previously trained for non-special characters but the model does not learn still to recognize the special characters.
I looked into some of the images from the ST_SPE and I was wondering if this is normal that some of them are very dark or my code somewhere messes up the channels.
Hello,
I have some issues with special characters...
Take a look : https://github.com/Belval/TextRecognitionDataGenerator/issues/243
Are you able to recognise these special characters?
Thanks you very much!
What is the ratio for ST_spe data?