From mipnerf360

Compute the exponential map `se(3) -> SE(3)`. This can be used for learning pose deltas on `SE(3)`. Args: tangent_vector: A tangent

Computes the expected value of sin(y) where y ~ N(x_means, x_vars) Args: x_means: Mean values. x_vars: Variance of values. R

@brief return flattened JSON value @sa https://json.nlohmann.me/api/basic_json/flatten/

Forward a zmq message to all websockets. Args: frames: byte messages to be sent over

Generate a point cloud from a nerf. Args: pipeline: Pipeline to evaluate with. num_points: Number of points to generate. May resu

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

grid_size: int of the grid height and width return: pos_embed: [grid_size*grid_size, embed_dim] or [n_cls_token+grid_size*grid_size, embe

Iterate over all the images in the eval dataset and get the average.

Returns the center of the box.

Retrieve the checkpoint directory

Returns the ImageDataset data as a dictionary. Args: image_idx: The image index in the dataset.

Returns the data for a specific image index. Args: image_idx: Camera image index

Get error bound from VolSDF

Calculates guassian approximation of conical frustum. Resturns: Conical frustums approximated by gaussian distribution.

Returns a 3 channel image. Args: image_idx: The image index in the dataset.

Returns a 3 channel image. Args: image_idx: The image index in the dataset.

Returns a 3 channel image. Args: image_idx: The image index in the dataset.

Method that can be used to process any additional metadata that may be part of the model inputs. Args: image_idx: The image index

Helper function to generate a placeholder frustum. Returns: A size 1 frustum with meaningless values.

Constructs and returns the OpenGL conversion matrix. Returns: numpy.ndarray: A 4x4 OpenGL conversion matrix.

Takes a camera path dictionary and returns a trajectory as a Camera instance. Args: camera_path: A dictionary of the camera path informat

Get a random color from a colormap. Args: colormap: List of colors. Defaults to Plotly colors. idx: Index of color to return. Def

Flattens RayBundle and extracts chunk given start and end indicies. Args: start_idx: Start index of RayBundle chunk.

Returns a list of camera in a sprial trajectory. Args: camera: The camera to start the spiral from. steps: The number of cam

Get a texture image.

Returns the training callbacks from both the Dataloader and the Model.

Create a new 'target_dir' + 'images' subfolder, and place user-uploaded images or extracted frames from video into it. Return (target_dir, im

! * @brief convert a byte to a uppercase hex representation * @param[in] byte byte to represent * @return representation ("00".."FF")

property: index

Draw some images and the scene aabb in the viewer. Args: dataset: dataset to render in the scene start_train: whether

return whether the iterator is at end

returns (is_good, err_msg)

@brief return whether value is an unsigned integer number @sa https://json.nlohmann.me/api/basic_json/is_number_unsigned/

Check if a viewpoint is valid and overlaps significantly with a reference one.

Checks if tensorboard is enabled.

Checks if wandb is enabled.

! @brief determine system byte order @return true if and only if system's byte order is little endian @note from https://stackoverflow.com/a/1001328

Load the checkpoint from the given path Args: loaded_state: dictionary of pre-trained model states

Convert binary mask to boundary mask. https://openaccess.thecvf.com/content/CVPR2021/papers/Cheng_Boundary_IoU_Improving_Object-Centric_Image

method: newTextureWithDescriptor:

Returns the next batch of data from the eval data manager. This will be a tuple of all the information that this data manager outputs.

Normalize the XYZs of poses to fit within a unit cube ([-1, 1]). Note: This operation is not in-place. Args: poses: A collection of poses

method: objectAtIndexedSubscript:

! @brief post-increment (it++) @pre The iterator is initialized; i.e. `m_object != nullptr`. */