r"""Function that measures Learned Perceptual Image Patch Similarity (LPIPS). Arguments: x, y (torch.Tensor): the input tensors to co

Dilate (via max-pooling) a non-negative step function.

Modify camera intrinsics to follow a different convention. Coordinates of the center of the top-left pixels are by default: - (0.5, 0.5)

Pad [..., 3, 4] pose matrices with a homogeneous bottom row [0,0,0,1].

tqdm but with parallel execution. Will essentially return res = [ function(arg) # default function(*arg) # if star_args is

Turn a PDF that integrates to 1 into a vector of weights that sums to 1.

Converts from (position, lookat, up) format to pose matrices.

Converts from pose matrices to (position, lookat, up) format.

Converts from pose matrices to (position, lookat, up) format.

extract 3D points/confidence from prediction head output

Convert quaternion to rotation in batch. Since all operation in pytorch, support gradient passing. Args: quad (tensor, batch_size*4)

Multiply two quaternions together using quaternion arithmetic

Return conjugate of quaternion. >>> q0 = random_quaternion() >>> q1 = quaternion_conjugate(q0) >>> q1[3] == q0[3] and all(q1[:3] == -q0[:

Return quaternion from rotation matrix. >>> R = rotation_matrix(0.123, (1, 2, 3)) >>> q = quaternion_from_matrix(R) >>> numpy.allclose(q,

Return multiplication of two quaternions. >>> q = quaternion_multiply([1, -2, 3, 4], [-5, 6, 7, 8]) >>> numpy.allclose(q, [-44, -14, 48, 28])

Return spherical linear interpolation between two quaternions. >>> q0 = random_quaternion() >>> q1 = random_quaternion() >>> q = quaterni

Return uniform random unit quaternion. rand: array like or None Three independent random variables that are uniformly distributed

see: src/base/reconstruction.cc void Reconstruction::ReadImagesBinary(const std::string& path) void Reconstruction::WriteImagesBi

Taken from https://github.com/colmap/colmap/blob/dev/scripts/python/read_write_model.py

Taken from https://github.com/colmap/colmap/blob/dev/scripts/python/read_write_model.py

see: src/base/reconstruction.cc void Reconstruction::ReadPoints3DBinary(const std::string& path) void Reconstruction::WritePoints

see: src/base/reconstruction.cc void Reconstruction::ReadPoints3DText(const std::string& path) void Reconstruction::WritePoints3D

Recenter poses around the origin.

Convert rotations given as rotation matrices to quaternions. Args: matrix: Rotation matrices as tensor of shape (..., 3, 3). Re

Compute rotation error Args: pose_error (4x4 array): relative pose error Returns: rot_error (float): rotation error

numpy version of sample()

Calculate n samples on a bspline cv : Array ov control vertices n : Number of samples to return degree: Curve d

Returns the skew-symmetric matrix of a vector cfo, 2015/08/13

Sparse alignment with MASt3R imgs: list of image paths cache_path: path where to dump temporary files (str) lr1, niter1: lea