About mutil camera

[lazyDamon]

Hello,
Do you have any thoughts adapting your work to multiple cameras ? When I use single camera intrinsics, everything is fine, but when I use multi-camera intrinsics, I get the following error:

File "MAtCha/2d-gaussian-splatting/train_with_charts.py", line 337, in training
    torch.log(1. + charts_scale_factor * (charts_depths[viewpoint_idx] - surf_depth).abs())
                                          ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~
RuntimeError: The size of tensor a (899) must match the size of tensor b (900) at non-singleton dimension 1