sudipansaha / dcvavhroptical Goto Github PK

Hello, @sudipansaha . Your method shows promising results and inspired me a lot. However, the layer combination (2,5,8) used in your code seems different from the setting (28,16,11,6) in your paper. Can you share the network structure in this code? Thank you very much.

On line 138 : state_dictForLayer23={k: v for k, v in netForFeatureExtractionLayer23Dict.items() if k in state_dictForLayer23}
netForFeatureExtractionLayer23Dict.items() seems to be a dictionary of randomly initialized parameters, while state_dictForLayer23 is the real pretrained_model. "for k, v in netF...items()" retains the weight in the random parameter.
So this line of code seems to just load random parameters into the network.
Nevertheless, thank you very much for migrating the code in matlab to the python version for us to learn.

Thanks for the repo. Do you have a sample code with the pre-trained models to test? Basically I want to pass two images and get what is changed between the two. Thanks.

Hi! Thanks for your great work. How can I download the Worldview-2 image pair, Pleiades image pair, and Quickbird image pair used in your paper?

Thank you.