buptxiaofeng

Hi, thanks for sharing this impressive work. The paper mentioned two architectures, Isotropic one and pyramid one. I noticed that in the code, this is a reduce_ratios, and this reduce_ratios...

how to deal with multi input

19

My CNN architecture is similar with the following figure. 

how to reproduce the results on the ICDAR2013 dataset reported in the paper?

1

I am trying to reproduce the reported scores on the ICDAR2013 dataset. Could you also release the processed ICDAR2013 dataset?

Computation resources and traning time

1

Could you provide some information about computation resources and traning time, if I want to train the model from scrath.

Hi, in the paper and the experimental results reported in the paper are Left Attribute and Top Attribute, while in the code the classes are defined as Header, Attribute. So...

https://github.com/Han-Jia/UNICORN-MAML/blob/c66b9f347f52afb5a988e4c679c9d398a41f9ae3/model/networks/convnet_maml.py#L31 Hi, thanks again for this impressive work. I really learned a lot. I have a question about the batchnorm in the implementation of convnet_maml. It seems that the variable...