Returns an iterator over all the names and meta parameters

Given convex volume1 and volume2, where volume1 is contained in volume2, this function returns the t0 at which rays leave volume1 and

Compute the KL divergence between two gaussians. Shapes are automatically broadcasted, so batches can be compared to scalars, among other

Partitions t0 and t1 into n_samples intervals. :param ts: [batch_size, *shape, n_samples, 1] :return: a tuple of (

Input: radius: local region radius nsample: max sample number in local region xyz: all points, [B, N, 3] new_xyz:

Sample a random subset of this PointCloud. :param num_points: maximum number of points to sample. :param subsample_kwargs: a

Registers a trainable or nontrainable parameter as a meta buffer. This can be later retrieved by meta_state_dict

Registers a meta parameter so it is included in named_meta_parameters and meta_state_dict.

Returns a backproppable rendering of a view :param batch: contains - height: Optional[int] - width: Optional

:param batch: contains either ["poses", "camera"], or ["cameras"]. Can optionally contain any of ["height", "width", "query_batch_size"]

Input: npoint: radius: nsample: xyz: input points position data, [B, N, 3] points: input points data, [B,

Input: xyz: input points position data, [B, N, 3] points: input points data, [B, N, D] Return: new_xyz: sampled point

Run farthest-point sampling on a batch of point clouds. :param points: batch of shape [N x num_points]. :param data_ctx: subsample count

Change every material to emit texture colors (or alpha) rather than having an actual reflective color. Returns a function to undo the changes

Calculate the spherical harmonics basis :param coords: [batch_size, *shape, 3] of unit norm :param sh_degree: Spherical harmonics degree

Calculate Euclid distance between each two points. src^T * dst = xn * xm + yn * ym + zn * zm; sum(src^2, dim=-1) = xn*xn + yn*yn + zn*zn

Write a PLY file for a mesh or a point cloud. :param coords: an [N x 3] array of floating point coordinates. :param rgb: an [N x 3] arra

:param outer_volume: is where distant objects are encoded.

Get the prior KL term for the variational lower-bound, measured in bits-per-dim. This term can't be optimized, as it only de

Like forward(), but always runs all encoders to ensure that the forward graph looks the same on every rank.

Create a light that attempts to match the light used by the Blender renderer when run with `--light_mode basic`.

:param toward: [batch_size x 3] unit vector from camera position to the object :param up: Optional [batch_size x 3] specifying the physical u

For every (x, y) coordinate in a rendered image, compute the ray of the corresponding pixel. :param coords: an [N x ... x 2]

Returns a [batch x num_views x 3*4+1] tensor of camera information.

Returns a [batch x num_views x 3*4+1] tensor of camera information.

Returns a [batch x num_views x 3*4+1] tensor of camera information.

Creates a new camera with the same intrinsics and direction as this one, but with a center crop to a square of the smaller dimension.

Compute the mean for the previous step, given a function cond_fn that computes the gradient of a conditional log probability with res

Sample x_{t-1} from the model using DDIM. Same usage as p_sample().

Use DDIM to sample from the model and yield intermediate samples from each timestep of DDIM. Same usage as p_sample_loop_pro

For every (x, y) coordinate in a rendered image, get the direction that corresponds to "depth" in an RGBD rendering. This ma

Returns a [batch x num_views x 1 x size x size] tensor in the range [-1, 1].

Returns a [batch x num_views x 1 x size x size] tensor in the range [-1, 1].

Compute the log-likelihood of a Gaussian distribution discretizing to a given image. :param x: the target images. It is assumed that this

Casts into OrderedDict and removes internal attributes

:return: [batch_size, num_views, n_patches, width]

Encode the batch into a per-data-point set of latents. :return: [batch_size, latent_ctx, latent_width]

Encode the batch into a single latent vector.

:return: [batch_size, num_views, n_patches, width]

:return: [batch_size, num_views, n_patches, width]