This function is borrowed from https://github.com/nkolot/SPIN/utils/geometry.py Find camera translation that brings 3D joints S closest to 2

Recursively find all files in the specified directory. Args: base (str): The base directory to start the search. Returns:

Recursively find all files in the specified directory. Args: base (str): The base directory to start the search. Returns:

Recursively find all files in the specified directory. Args: base (str): The base directory to start the search. Returns:

Recursively find all files in the specified directory. Args: base (str): The base directory to start the search. Returns:

Input: quat_params:(seq, j, 4) local quternion rotation of each joint root_pos: (seq, 3), root global postion after r

Computes continuous 6D parameters, root linear velocity, and root angular velocity based on the input motion capture data. A

Load yaml config files from subfolders

Adjusts the motion capture data to a local pose representation and ensures that all poses face in the Z+ direction. Args:

Convert the loss name to a log name.

Convert rotations given as rotation matrices to axis/angle. Args: matrix: Rotation matrices as tensor of shape (..., 3, 3). Ret

https://github.com/duolu/pyrotation/blob/master/pyrotation/pyrotation.py Convert a rotation matrix to a unit quaternion. This uses the Sh

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

Converts rotation matrices to 6D rotation representation by Zhou et al. [1] by dropping the last row. Note that 6D representation is not uniq

Parse arguments and load config files

Plot 3D motion data. Parameters: - joints (numpy array): 3D motion data of joints, shape (frame_number, joint_number, 3). - out_name

Plot 3D motion data of SMPL-H model. Parameters: - joints (numpy array): 3D motion data of joints, shape (frame_number, joint_number, 3)

Processes motion capture data, including downsampling, skeleton normalization, floor alignment, and feature extraction. Args: po

find the quaternion used to rotate v0 to v1

Enforce quaternion continuity across the time dimension by selecting the representation (q or -q) with minimal distance (or, equivalently, ma

q0 : tensor of quaternions t: tensor of powers

This function is borrowed from https://github.com/MandyMo/pytorch_HMR/blob/master/src/util.py#L50 Convert quaternion coefficients to rotatio

Given a quaternion representing rotation, get the quaternion representing its inverse. Args: quaternion: Quaternions as tensor o

Given a quaternion representing rotation, get the quaternion representing its inverse. Args: quaternion: Quaternions as tensor o

This function is borrowed from https://github.com/kornia/kornia Convert quaternion vector to angle axis of rotation. Adapted from ceres

Convert rotations given as quaternions to axis/angle. Args: quaternions: quaternions with real part first, as tensor of

Convert rotations given as quaternions to axis/angle. Args: quaternions: quaternions with real part first, as tensor of

Convert rotations given as quaternions to rotation matrices. Args: quaternions: quaternions with real part first, as ten

Generate random quaternions representing rotations, i.e. versors with nonnegative real part. Args: n: Number of quaternions in a

Generate random quaternions representing rotations, i.e. versors with nonnegative real part. Args: n: Number of quaternions in a

Generate random rotations as 3x3 rotation matrices. Args: n: Number of rotation matrices in a batch to return. dtype: Type t

Generate random rotations as 3x3 rotation matrices. Args: n: Number of rotation matrices in a batch to return. dtype: Type t