:param face_idxs: [cam_num,res,res] per pixel face id :return face_view_pixel_num:

:param foreground_mask: res,res :param method: 'shift' or 'dilate'. 'dilate' produces more smooth_depth_map result :return:

Get minimum squared distance between point and cube. Params: point_coord : cartesian coordinates of point no_dims : number of dimensions b

Get minimum squared distance between point and cube. Params: point_coord : cartesian coordinates of point no_dims : number of dimensions b

:param center: [3] :param vector: [3] :param color: :return:

:param point_pixels: :param colors: :param point_validation: :param hard_mask: :return:

Get a search algorithm by name. # Arguments search_name: string. The name of the search algorithm.

This matches the implementation in Denoising Diffusion Probabilistic Models: From Fairseq. Build sinusoidal embeddings. This matches

DON'T USE THIS, EVEN SLOWER :param masked_imgs: torch tensor, [N,3,H,W](RGB), float, 0-1. H=W=256 :param masks: torch tensor, [N,H,W], float3

CHW,RGB, float, 0-1 :param file_name: :return:

load an ``obj`` mesh. Args: path (str): path to mesh. albedo_path (str, optional): path to the albedo texture image,

load a mesh using ``trimesh.load()``. Can load various formats like ``glb`` and serves as a fallback. Note: We will try

Log a value of some diagnostic Call this once for each diagnostic quantity, each iteration If called many times, last value will be used.

:param min: min coordinates of the grid :param max: :param resolution: :return: p: a torch tensor on cpu of shape: resolution^3, 3

Makes a 3D grid. Args: bb_min (tuple): bounding box minimum bb_max (tuple): bounding box maximum shape (tuple): output s

Makes a 3D grid. Args: bb_min (tuple): bounding box minimum bb_max (tuple): bounding box maximum shape (tuple): output sh

* Returns a pointer to a piece of new memory of the given size in bytes * allocated from the pool. */

Copy the master parameter data back into the model parameters.

Defaults for image training.

Copy the gradients from the model parameters into the master parameters from make_master_params().

Copied from: https://github.com/openai/baselines/blob/ea25b9e8b234e6ee1bca43083f8f3cf974143998/baselines/common/mpi_util.py#L110 Perform a we

Numpy :param img: C,H,W :param mask2: H,W. The background is True, foreground non-empty is True, foreground empty is False (to be painted

Find the nearest (Euclidean) neighbor in dst for each point in src Input: src: Nxm array of points dst: Nxm array of points

input_rgb: [B, N,3]. from 0-1

Partition a range of data points by manipulation the permutation index. The sliding midpoint rule is used for the partitioning. Params: pa : data

Partition a range of data points by manipulation the permutation index. The sliding midpoint rule is used for the partitioning. Params: pa : data

:param points: N,3 :param mesh_verts: :param mesh_faces: :return: N,3

:param vertices: [V,3] :param faces: [F,3] :param images: [img_num, 3,res,res] :param face_view_ids: [F] :param face_vertex_uvs:

read XYZ point cloud from filename PLY file

:param mesh_data: : a KaolinDatasetItem. Can be obtained by kal.io.shapenet.ShapeNetV2()[i]. mesh_data.data: a kaolin.io.obj.ObjMesh

:param coords: :param colors: float within range of 0-1 :param path: :return:

:param coords: :param colors: float within range of 0-1 :param path: :return: