About CameraFusion dataset

[xianrui-luo]

Hi, thank you for your code and dataset. I wonder how you collect the corresponding telephotos for wide photos? Did you manually zoom in or you have other techniques? If it is from zooming in, what is the zoom-in ratio?

[Brightlcz]

Hi,
you may refer to #1.

[xianrui-luo]

Hi, you may refer to #1.

Thank you for your reply. I would like to ask one more questions about the parameter "num_patch". It is set to 300, which is really large, and it takes a long time to train. Is it right to do so?

[jiaxinxie97]

Hi,
we set "num_patch" to 300, so in each epoch we already sample every training image 300 times, it is large, so you don‘t need to run default 1000 epochs, 100 epochs is enough and you can use val dataset to verify if it can stop.

[xianrui-luo]

In my dataset, 100 epochs probably need a whole week to train? Is this the normal case?