Change the camera parameters, but get the same output

[visonpon]

Hi @griegler, When I try to modify the Ks.npy/Rs.npy/ts.npy for synthesis new view images, the output results are always the same as the original camera parameters' output, hope you can help, thanks~

[griegler]

Can you give a couple more details what you try to achieve and what you tried? I am not sure I fully understand the problem.

[visonpon]

I'm sorry that I didn't make it clear, what I want to do is to synthesis a new view image, so I try to modify the camera's parameters and hope that by giving new camera parameters the model can output a new synthesized image that never existed before. But after change the camera parameters, the output didn't have any change
AM I have some misunderstanding about the paper?

[griegler]

Note that you also need the depth map in the target view and the overlap to the source views.
You could have a look at the code here. It is the script I used to create custom tracks.

[visonpon]

yeah, the target view is what I want to synthesize, but how to get its depth map before the target view image synthesized since it's a new view that hasn't come out before

[griegler]

You render the depth map of the target view from the proxy geometry (mesh reconstructed from the source images).

[visonpon]

oh, I got it, Thanks for your patient response~