Sample blurred image Ground truth

I want to deblur text images using deep learning. Which approaches are best suited for the task? Any example networks? Is unsupervised network the best approach? GAN or cycle GAN for these purposes?

I have currently prepared 1000 images for training (shapr+blur) is it sufficient? For each of these approaches, how many training images do I need? I have attached sample blurred image and Ground truth

  • $\begingroup$ Are these purposefully blurred pictures or shaky camera work? $\endgroup$ – Lio Elbammalf Sep 9 '19 at 8:48
  • $\begingroup$ @LioElbammalf the blurred images were creating by applying motion blur filter to simulate real camera like distortion in blur images. $\endgroup$ – Bot_Start Sep 9 '19 at 9:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.