Encode images with backbone and decode into a semantic segmentation map of the same size as input.

Placeholder of forward function.

Gather all CSV files in the input directory, concatenate them, and write to the output file. If the input directory contains multiple rows in

embed_dim: output dimension for each position pos: a list of positions to be encoded: size (M,) out: (M, D)

embed_dim: output dimension for each position pos: a list of positions to be encoded: size (M,) out: (M, D)

This function generates a 1D positional embedding from a given grid using sine and cosine functions. Args: - embed_dim: The embedding di

embed_dim: output dimension for each position pos: a list of positions to be encoded: size (M,) out: (M, D)

this will perform the index select, cat the tensors, and provide the attn_bias from cache

this will perform the index select, cat the tensors, and provide the attn_bias from cache

this will perform the index select, cat the tensors, and provide the attn_bias from cache

this will perform the index select, cat the tensors, and provide the attn_bias from cache

this will perform the index select, cat the tensors, and provide the attn_bias from cache

this will perform the index select, cat the tensors, and provide the attn_bias from cache

Create a composition of image augmentations. Args: color_jitter: Dictionary containing color jitter parameters: - brightness:

Args: rays_o: (..., n_rays, 3) ray origins rays_d: (..., n_rays, 3) ray directions. z_vals: (..., n_rays, n_samples) z va

Return set of symbol pairs in a word. Word is represented as tuple of symbols (symbols being variable-length strings).

poses: list of 4x4 arrays

Get image space UV grid, ranging in [0, 1]. >>> image_uv(10, 10): [[[0.05, 0.05], [0.15, 0.05], ..., [0.95, 0.05]], [[0.05, 0.15],

Get OpenCV intrinsics matrix Args: focal_x (float | torch.Tensor): focal length in x axis focal_y (float | torch.Tensor): fo

Linear interpolation between two vectors. Args: x1 (np.ndarray): [..., d] vector 1 x2 (np.ndarray): [..., d] vector 2

Loads images from a directory or video, resizes them to a uniform size, then converts and stacks them into a single [N, 3, H, W] PyTorch tens

Read trajectory format. Return in TUM-RGBD format. Returns: traj_tum (N, 7): camera to world poses in (x,y,z,qx,qy,qz,qw) timestam

Make the input directory for CSV files and remove any existing history CSV files.

This function generates a 1D positional embedding from a given grid using sine and cosine functions. Args: - embed_dim: The embedding di

Convert a batch of 3x3 rotation matrices to axis-angle representation (rotation vector) Args: rot_mat (torch.Tensor): shape (..., 3, 3),

Convert 3x3 rotation matrix to quaternion (w, x, y, z) Args: rot_mat (np.ndarray): shape (..., 3, 3), the rotation matrices to convert

MipNeRF rendering. Args: mipnerf: mipnerf model, which takes (points_mu, points_sigma) as input and returns (color, density) as outp

NeRF rendering of rays. Note that it supports arbitrary batch dimensions (denoted as `...`) Args: nerf: nerf model, which takes (poi

Given color, sdf values and z_vals (linear depth of the sampling points), do volume rendering. (NeuS) Args: color: (..., n_samples o

UV with left-top corner as (-width / diagonal, -height / diagonal) and right-bottom corner as (width / diagonal, height / diagonal)

Get OpenGL perspective matrix Args: fov_y (float | torch.Tensor): field of view in y axis aspect (float | torch.Tensor): asp

Get OpenGL perspective matrix Args: fov_y (float | np.ndarray): field of view in y axis aspect (float | np.ndarray): aspect

nested_split_lens: A list of N lists of ints. Each int indicates the length of a split within a sample, where each sample contains multi

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

Rasterize a mesh with vertex attributes. Args: ctx (GLContext): rasterizer context vertices (np.ndarray): (B, N, 2 or 3 or 4

Rasterize vertex attribute. Args: vertices (np.ndarray): [N, 3] faces (np.ndarray): [T, 3] attr (np.ndarray): [N, C]

Reads a trajectory from a text file. File format: The file format is "stamp d1 d2 d3 ...", where stamp denotes the time stamp (to b

Rotates half the hidden dims of the input.

Rotates half the hidden dims of the input.

Rotates half the hidden dims of the input.

Rotates half the hidden dims of the input.

r"""Sample spatial features `sample_features4d(input, coords)` samples the spatial features :attr:`input` represented by a 4D tensor :math:`(

images: np.ndarray of shape (N, H, W, 3) in [0, 255] or torch.Tensor of shape (N, 3, H, W) in range [0, 1]

Keeps only the latest 'keep_latest' checkpoints in ckpt_dir. Assumes checkpoint folders are named with step numbers like 0001000/.

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

Processes specialized tokens with shape (1, 2, X, C) for multi-frame processing: 1) Uses the first position (index=0) for the first frame onl

Return x view of the input array with x sliding window of the given kernel size and stride. The sliding window is performed over the given ax

Rescales the image so that the following conditions are met: 1. Both dimensions (height and width) are divisible by 'factor'. 2. The total n

Solve `min |focal * xy / (z + shift) - uv|` with respect to shift and focal

Solve `min |focal * xy / (z + shift) - uv|` with respect to shift

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