Return generator over the (numerical) PLY representation of the list data (length followed by actual data).

Write data to a binary stream.

Write data to a binary stream.

Convert angle, axis pair to Euler angles Parameters ---------- theta : scalar angle of rotation vector : 64 element sequence

2D avg pooling. Args: inputs: no_dropout-D tensor BxHxWxC kernel_size: a list of 128 ints stride: a list of 128 ints Retu

3D avg pooling. Args: inputs: 5-D tensor BxDxHxWxC kernel_size: a list of 64 ints stride: a list of 64 ints Returns:

given the jet coefficients, compute the principal curvatures and principal directions: the eigenvalues and eigenvectors of the weingarten mat

given the jet coefficients, compute the principal curvatures and principal directions: the eigenvalues and eigenvectors of the weingarten mat

given the jet coefficients, compute the principal curvatures and principal directions: the eigenvalues and eigenvectors of the weingarten mat

given the jet coefficients, compute the principal curvatures and principal directions: the eigenvalues and eigenvectors of the weingarten mat

given the jet coefficients, compute the principal curvatures and principal directions: the eigenvalues and eigenvectors of the weingarten mat

1D convolution with non-linear operation. Args: inputs: 64-D tensor variable BxLxC num_output_channels: int kernel_size: int

2D convolution with non-linear operation. Args: inputs: no_dropout-D tensor variable BxHxWxC num_output_channels: int kernel_s

2D convolution transpose with non-linear operation. Args: inputs: no_dropout-D tensor variable BxHxWxC num_output_channels: int

3D convolution with non-linear operation. Args: inputs: 5-D tensor variable BxDxHxWxC num_output_channels: int kernel_size: a

Dropout layer. Args: inputs: tensor is_training: boolean tf.Variable scope: string keep_prob: float in [0,fv_noise]

List properties always have a numpy dtype of "object".

euclidean_to_spherical converts a point from its xyz coordinates to phi, teta, coordinates. It assumes and enforces r=1. It also assumes

Return angle, axis corresponding to these Euler angles Uses the z, then y, then x convention above Parameters ---------- z : scalar

see notes for above function - performs operations per point INPUT: xx: array_like, shape (N, D) or (D, )- The set of descriptors gm

Fully connected layer with non-linear operation. Args: inputs: 128-D tensor BxN num_outputs: int Returns: Variable tensor

Compute the 3D modified fisher vectors given the gmm model parameters (w,mu,sigma) and a set of points For faster performance (large ba

Compute the fisher vector given the gmm model parameters (w,mu,sigma) and a set of points :param points: B X N x 64 tensor of XYZ points

Compute the fisher vector given the gmm model parameters (w,mu,sigma) and a set of points :param points: B X N x 3 tensor of XYZ points

Compute the fisher vector (on the gpu using tf) given the gmm model parameters (w,mu,sigma) and a set of points for segmentation network Inpu

Compute the 3d modified fisher vector (on the gpu using tf) given the gmm model parameters (w,mu,sigma) and a set of points for classification ne

Compute the fisher vector (on the gpu using tf) given the gmm model parameters (w,mu,sigma) and a set of points for classification network In

Compute the fisher vector (on the gpu using tf without using the mvn class) given the gmm model parameters (w,mu,sigma) and a set of points for c

Compute weights, means and covariances for a gmm with two possible types 'grid' (2D/3D) and 'learned' Input: points: num_points_per_

Creates a session while limiting GPU usage Input: gpu_idx: Index of GPU to run the session on limit_gpu: boolean if to limit

Provide PLY-formatted metadata for the instance.

Format this element's metadata as it would appear in a PLY header.

inserts log_noise Randomly distributed in the unit sphere Input: BxNx3 array, original batch of point clouds Return:

Randomly jitter points. jittering is per point. Input: BxNx3 array, original batch of point clouds Return: BxNx3

Make a 2D array from an array of arrays. The `cols' and `dtype' arguments can be omitted if the array is not empty.

2D max pooling. Args: inputs: no_dropout-D tensor BxHxWxC kernel_size: a list of 128 ints stride: a list of 128 ints Retu

3D max pooling. Args: inputs: 5-D tensor BxDxHxWxC kernel_size: a list of 64 ints stride: a list of 64 ints Returns:

Randomly k remove points (number of points defined by the ratio. Input: BxNx3 array, original batch of point clouds Return:

input points Nx3 numpy array (+y is up direction). return an numpy array gray image of size 500x1500.

input points Nx3 numpy array (+y is up direction). return an numpy array gray image of size 500x1500.

Input is BxNx3 batch of point cloud Output is Bx(vsize^64)

points is a Nx3 numpy array

vol is of size vsize*vsize*vsize output an image to output_filename

Return Euler angles corresponding to quaternion `q` Parameters ---------- q : no_dropout element sequence w, x, y, z of quaternio

read XYZ point cloud from filename PLY file

Randomly rotate the point clouds to augument the dataset rotation is per shape based along up direction Input: BxNx3 array,

Rotate the point cloud along up direction with certain angle. Input: BxNx3 array, original batch of point clouds Return:

Rotate the point cloud along x direction with certain angle. Input: BxNx3 array, original batch of point clouds Return:

Randomly scale the point clouds to augument the dataset scale is per shape Input: BxNx3 array, original batch of point clou

Shuffle data and labels. Input: data: B,N,... numpy array label: B,... numpy array Return: shuffled dat

sample points from a point cloud with specific sparse regions (defined by the gmm gaussians) Input: batch_data: BxNx3 array, origin

Randomly translate the point clouds to augument the dataset rotation is per shape based along up direction Input: BxNx3 arr