Open guwinston opened 2 months ago
Hi, currently we do not support training with different image resolutions and camera intrinsics. But theoretically, our distribution solution should support both them. I think it is not complicated to implement them in our repository. You can have a try!
If you really want to train on different resolutions, we can schedule a meeting and talk about your need, then I can help add this functionality.
Does this repository support training datasets with different image resolutions and intrinsic parameters like the original 3DGS repo