How can I generate or obtain the visualization result for predicted characters?

[mhliao]

How can I generate or obtain the visualization result for predicted characters?

[MhLiao]

@Peony-Wang Once you get the detection results of our method, you can crop the text regions for recognition (eg. CRNN or ASTER).

[MhLiao]

@Peony-Wang What's the meaning of predicted characters? Do you mean the recognition results? This repo is only a scene text detector which produces bounding boxes of the text instances.

[Peony-Wang]

@Peony-Wang What's the meaning of predicted characters? Do you mean the recognition results? This repo is only a scene text detector which produces bounding boxes of the text instances.

Got it. Thanks for your reply! I have one more question, could you give a clue bout how to combine your work with other detection+recognition work to improve the result of the recognition result?

[Peony-Wang]

Thank you for your answer！