Open vijay1131 opened 3 years ago
I do have the same problem.
I hope you do not mean the GAN recognizer, as we intentionally made it sub-optimal (see details in the paper and the supplementary materials). We use the clova text recognition benchmark for all the experiments almost as-is (we just added affine augmentations during training).
Finally, note that the biggest boost we got was for domain adaptation, YMMV.
Hope this helps
HI I would like to know how to proceed for testing the recogniser model to evaluate the text recognition. Also would that be a good idea to use recogniser model from this project?
Thanks a lot !