Christian Bartz
Christian Bartz
Hi, you should be able to use this network to perform rectification of already cropped scene text with this model. But this model works in such a way, that it...
I already shared the model, please have a look at [this website](https://bartzi.de/research/see), and also read the correspinding section of the [README](https://github.com/Bartzi/see#pretrained-models) :wink:
Sure :wink: Just find the necesary training data and use create a new `char_map` only with the ascii codes for captial alphanumerics. Train and enjoy, yhis should be it...
Hi, good question. I never really cared about image resolution or quality, I mean the text should be readable, but that is about it. All samples that you put into...
No, there is no need for bounding boxes! This is the point of the paper, you only need an image and the textual content as annotation. The README and the...
I'm sorry but I don't see your problem. If you want to retrain, or finetune, you just have to prepare everything as if you'd like to train the model from...
That is kind of right. I added these loss_weights in order to encourage the network to correct errors in the third word, since the dataset mainly consists of images with...
Which pre-trained model are you using? Are you using the one from `bartzi.de`?
Did you have a look at the `bboxes` directory that has been created in the log directory for this train run? How do these images look like (there should be...
Oh now I see the problem^^ Try to use `-r ../model/model_190000.npz` instead of `--model ../model/model_190000.npz` It should work then