Can we use Yolo to detect and recognize text in an image?

Question

I am currently using a deep left model called "Yolov2" to detect an object, and I want to use it to extract text and use it to save it to disk, but I don’t know how to do it if anyone finds out more about this, please advise me

I use Tensorflow

thanks

Answer 1

If you are using a pre-trained model, you need to save this output and enter the images into the character recognition network if a neural network or other approach is used.

What you do is "scene text recognition". You can check the reading of text in the wild using convolutional neural networks , paper , here a demo and the home page . Github chongyangtao has a list of related resources.

Answer 2

I have a similar question and am making a model for detecting digits using the svhn dataset. This is not a finished project, but it works well. You can see the Yolo-digit-detector code.