3j SSKJr SSKrSSKJr SSKrSSKJr S S SjjrSSjr SSjr SSjr SS jr SS jr SS jrg)) annotationsN)Union) PinholeCamerac [R"[R"S/U-XS9[R"S/U-XS945n[R"[R"S/U-XS9[R"S/U-XS945n[R"[R"S/U-XS9[R"S/U-XS945nXEU4$)a@Create camera dimensions for ray sampling. Args: device: Union[str, torch.device, None] for tensors dtype: Data type for tensors n_cams1: Number of cameras in first group (default: 3) n_cams2: Number of cameras in second group (default: 2) Returns: Tuple of (heights, widths, num_img_rays) tensors devicedtypedi,i )torchcattensor)r r n_cams1n_cams2heightswidths num_img_rayss V/home/wildlama/miniconda3/lib/python3.13/site-packages/kornia/geometry/camera/utils.pycreate_camera_dimensionsrs"II LL# E LL# E G !99 LL# E LL# E F "' LL" D LL" D "L L ((c/n[U[R5(aUR5OUn[U[R5(aUR5OUn[ XXx5H<upp[R "SXES9n XS'XS'XS'XS'UR U 5 M> [R"U5$)a{Create intrinsic camera matrices from focal lengths and principal points. Args: fxs: Focal length in x direction fys: Focal length in y direction cxs: Principal point x coordinate cys: Principal point y coordinate device: Union[str, torch.device, None] for tensors dtype: Data type for tensors Returns: Stacked intrinsic matrices of shape (N, 4, 4) rrrrrrr) isinstancerTensortolistzipeyeappendstack)fxsfyscxscysr r intrinsics_batchcxs_listcys_listfxfycxcy intrinsicss rcreate_intrinsicsr4@s*,.)#u||<rtr?rtr@rtr rnr rorprs) rUfloatrVrur rnr rorpr)r rnr rorpr)r]rrplist[torch.Tensor])r]rr rnrprv) __future__rrStypingrrkornia.geometry.camerarrr4rNrWr^rirmrrr{s/$#  1bc ) * )3> )IL )[^ )4 )F )  )  ) * ) * ) + )  ) )F7) 7) 7) 7)  7)  7)  7) +7) 7)7)t# ##)G#P[##L Fr