2x2 matrix for 2D rotation Args: theta (float | torch.Tensor): rotation angle in radians, arbitrary shape (...,) Returns:

images: numpy array of shape (N, H, W, 3) grid_shape: (rows, cols)

Saves a PLY visualization of the world points and images. Args:

Forward pass through the fusion blocks. Args: features (List[Tensor]): List of feature maps from different layers.

Convert rotation matrix and translation vector to 4x4 transformation matrix. Args: R (np.ndarray): [..., 3, 3] rotation matrix

Placeholder for single image test.

Read camera data, return (M,N) tuple. M is the intrinsic matrix, N is the extrinsic matrix, so that x = M*N*X, with x being a point in h

Skew symmetric matrix from a 3D vector

Skew symmetric matrix from a 3D vector

Spherical linear interpolation between two rotation matrices Args: rot_mat_1 (torch.Tensor): shape (..., 3, 3), the first rotation matrix

Spherical linear interpolation between two unit quaternions. Args: q1 (np.ndarray): [..., d] unit vector 1 q2 (np.ndarray):

Spherical linear interpolation between two rotation matrices. Args: R1 (np.ndarray): [..., 3, 3] rotation matrix 1 R2 (np.nd

Sliding window view of the input tensor. The dimension of the sliding window is appended to the end of the input tensor's shape. NOTE: Since

Solve the quad mesh from the candidate quad faces. Args: face2edge (np.ndarray): [T, 3] face to edge relation edge2face (np.

Convert a unit quaternion to a standard form: one in which the real part is non negative. Args: quaternions: Quaternions with re

Convert 9D representation to SO(3) using SVD orthogonalization. Args: m: [BATCH, 3, 3] 3x3 matrices. Returns: [B

Convert 9D representation to SO(3) using SVD orthogonalization. Args: m: [BATCH, 3, 3] 3x3 matrices. Returns: [B

Create sinusoidal timestep embeddings. :param t: a 1-D Tensor of N indices, one per batch element. These ma

Transfer some variables to another device (i.e. GPU, CPU:torch, CPU:numpy). batch: list, tuple, dict of tensors or other things device: pytor

Iteratively refine camera pose predictions. Args: pose_tokens (torch.Tensor): Normalized camera tokens with shape [B, 1,

Estimates the Sim(3) transformation between `X` and `Y` point sets. Estimates c, R and t such as c * R @ X + t ~ Y. Parameters ----

Update the .vscode/launch.json file with the new host and port.

Update the .vscode/launch.json file with the new host and port.

Warp kpts0 from I0 to I1 with depth, K and Rt Also check covisibility and depth consistency. Depth is consistent if relative error < 0.2 (hard

Warp kpts0 from I0 to I1 with depth, K and Rt Also check covisibility and depth consistency. Depth is consistent if relative error < 0.2 (hard

Inference with full image.

This method defines the job that the node should do for each input item. A item obtained from the input queue is passed as arguments

Main method to prepare for the model one or several sequences(s) and image(s). This method forwards the `text` and `kwargs` arguments

Submit data to the pipeline's input queue, and return the output data asynchronously. NOTE: The pipeline must be streamed (i.e., ever

Generates spatial positions for a batch of patches. Args: batch_size: Number of samples in the batch. height: Height

Fetch item by index and a dynamic variable n_per_seq.

Fetch item by index and a dynamic variable n_per_seq.

Fetch item by index and a dynamic variable n_per_seq.

Fetch item by index and a dynamic variable n_per_seq.

Fetch item by index and a dynamic variable n_per_seq.

Fetch item by index and a dynamic variable n_per_seq.