Open pzz2011 opened 7 years ago
@pzz2011 Yes. Answer in paper.
For all the experiments for scene text recognition, we use the synthetic dataset (Synth) released by Jaderberg et al. [20] as the training data.
[20] M. Jaderberg, K. Simonyan, A. Vedaldi, and A. Zisserman. Synthetic data and artificial neural networks for natural scene text recognition. NIPS Deep Learning Workshop, 2014. 5
What are the steps to retrain the model from the synthetic dataset? For example, what is passed in as the parameters for createDataset.py?
Hi, Bgshih,
What's dataset the pretrained model use? Is it http://www.robots.ox.ac.uk/~vgg/data/text/mjsynth.tar.gz ?[this file is too big....] I'm confused about it.
If you have time, I hope I can receive you reply . thanks!