Open ghost opened 6 years ago
First of all, Sorry for my ignorance. Since I'm new in the IT industry, especially Deep Learning stuff.
I think EAST is easy to train and works pretty well(both image and video).
But I guess a lot of people(including me)'s goal is not only text detection itself but also text recognition(convert to image to real text).
I try to use tesseract with EAST in using webcam. But It doesn't work well.
I don't know how to combine text-detection with text-recognition
So can you tell me some advice or some resources related to?
normally, firstly using east to crop the word area then the using other recognition frameworks (like crnn) to get the text content from the cropped parts is the way how most companies do.
First of all, Sorry for my ignorance. Since I'm new in the IT industry, especially Deep Learning stuff.
I think EAST is easy to train and works pretty well(both image and video).
But I guess a lot of people(including me)'s goal is not only text detection itself but also text recognition(convert to image to real text).
I try to use tesseract with EAST in using webcam. But It doesn't work well.
I don't know how to combine text-detection with text-recognition
So can you tell me some advice or some resources related to?