Apply joint transforms with priority: triple transforms > dual transforms.

Extended version that handles pointmaps

Apply transformation to Gaussian Splatting object IN-PLACE for efficiency. Returns backup data to restore later.

Args: quaternion: (4,) tensor of quaternion translation: (3,) tensor of translation

R3 camera space --> PyTorch3D camera space Also needed for pointmaps

Main function to check different types of occlusion for a given mask and 3D point map.

Decode the structured latent. Args: slat (sp.SparseTensor): The structured latent. formats (List[str]): The

Returns: scale: (..., 3) tensor of scale factors rotation: (..., 3, 3) tensor of rotation matrices translation: (..., 3)

Extract a mask from either the alpha channel of an RGB image or a depth image. Args: rgb_image: Tensor of shape (B, C, H, W) or (C,

Main function to visualize Cameras, Meshes, Pointclouds, and RayBundle. Plots input Cameras, Meshes, Pointclouds, and RayBundle data into nam

Render the gausssian. Args: gaussian : gaussianmodule extrinsics (torch.Tensor): (4, 4) camera extrinsics

Broadcast a 1D tensor to a sparse tensor along the batch dimension then perform an operation. Args: input (torch.Tensor): 1D tensor

Apply scaled dot product attention to a sparse tensor. Args: qkv (SparseTensor): A [N, *, 3, H, C] sparse tensor containing Qs, Ks,

Split a 4-channel RGBA image into a 3-channel RGB image and a 1-channel mask. Args: rgba_image: A 4-channel RGBA image. Returns

Adds a trace rendering a Meshes object to the passed in figure, with a given name and in a specific subplot. Args: fig: plotly f

Build the probability distribution for dropout configurations. Returns: dropout_configs: List of sets containing

Regularizer L_dev as in Equation 8

Computes the location of dual vertices as described in Section 4.2

Creates a default camera trajectory, rotating around the scene in a circle. Args: num_frames: Number of frames in the trajectory

Rasterize a mesh from multiple views and remove invisible faces. Also includes postprocessing to: 1. Remove connected components that

Takes in the number of objects to be plotted and generate a plotly figure with the appropriate number and orientation of titled subplots.

Obtains the ID of topology cases based on cell corner occupancy. This function resolves the ambiguity in the Dual Marching Cubes (DM

Identifies grid cubes that intersect with the underlying surface by checking if the signs at all corners are not identical.

Identifies grid edges that intersect with the underlying surface by checking for opposite signs. As each edge can be shared by multi