I wanted to know if there is a way to extract hardcoded subtitles via OCR, should I do some image processing after extracting the frames in order to use tesseract afterwards?

I have tried to extract frames from the video and remove everything that is not the color of the subtitles, but I got terrible results when using tesseract afterwards.

Is it possible to get decent results only using some workaround like this or should I rather train tesseract myself? It is easy to generate training data from random quotes added on random movie pictures with random fonts but it will require a lot of time to train it.

  • 3
    It may help to show us an image after the color is removed. – agc Jul 17 '18 at 23:55

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.