Helper to create a Variable stored on CPU memory. Args: name: name of the variable shape: list of ints initializer: initializer for Vari

Input: points: (batch_size, ndataset, channel) float32 array, points to sample from idx: (batch_size, npoint, nsample) int32 arra

input: batch_size * ndataset * 3 float32 batch_size * npoints int32 returns: batch_size * npoints * 3 float32

Helper to create an initialized Variable with weight decay. Note that the Variable is initialized with a truncated normal distribution. A weight

Batch normalization on convolutional maps and beyond... Ref.: http://stackoverflow.com/questions/33949786/how-could-i-use-batch-normalization-in-te

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

Input: radius: float32, ball search radius nsample: int32, number of points selected in each ball region xyz1: (batch_siz

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

Input: points: (b,m,c) float32 array, known points idx: (b,n,3) int32 array, indices to known points weight: (b,n,3) floa

Input: xyz1: (b,n,3) float32 array, unknown points xyz2: (b,m,3) float32 array, known points Output: dist: (b,n,3) fl

input: int32 batch_size * ndataset * 3 float32 returns: batch_size * npoint int32

PointNetVLAD, INPUT is batch_num_queries X num_pointclouds_per_query X num_points_per_pointcloud X 3, OUTPUT batch_num_que

Batch normalization on 2D convolutional maps. Args: inputs: Tensor, 4D BHWC input maps is_training: boolean tf.Varialbe, true i

Input: k: int32, number of k in k-nn search xyz1: (batch_size, ndataset, c) float32 array, input points xyz2: (batch_size

PointNet Feature Propogation (FP) Module Input:

Batch normalization on 1D convolutional maps. Args: inputs: Tensor, 3D BLC input maps is_training: boolean tf.Varialbe, true in

Batch normalization on 3D convolutional maps. Args: inputs: Tensor, 5D BDHWC input maps is_training: boolean tf.Varialbe, true

Batch normalization on FC data. Args: inputs: Tensor, 2D BxC input is_training: boolean tf.Varialbe, true indicates training ph

Context Gating Args: input_layer: Input layer in the following shape: 'batch_size' x 'number_of_activation' Returns: activation:

Feature Transform Net, input is BxNx1xK Return: Transformation matrix of size KxK

CONSTANT WEIGHT TODO input: dist (b,n,3) output: weight (b,n,3)

input: points (b,n,c), idx (b,m,nsample) output: out (b,m,nsample,c)

input: grad_out (b,m,nsample,c), idx (b,m,nsample), output: grad_points (b,n,c)

Input (XYZ) Transform Net, input is BxNx3 gray image Return: Transformation matrix of size 3xK

input: points (b,m,c), idx (b,n,3), weight (b,n,3) output: out (b,n,c)

input: grad_out (b,n,c), idx (b,n,3), weight (b,n,3) output: grad_points (b,m,c)

PointNet Set Abstraction (SA) Module Input: xyz: (batch_size, ndataset, 3) TF tensor points: (batch_size, ndataset, c

PointNet Set Abstraction (SA) module with Multi-Scale Grouping (MSG) Input: xyz: (batch_size, ndataset, 3) TF tensor

input: batch_size * ncategory float32 batch_size * npoints float32 returns: batch_size * npoints int32

input: radius (1), nsample (1), xyz1 (b,n,3), xyz2 (b,m,3) output: idx (b,m,nsample)

Input: npoint: int32 radius: float32 nsample: int32 xyz: (batch_size, ndataset, 3) TF tensor points: (bat

Inputs: xyz: (batch_size, ndataset, 3) TF tensor points: (batch_size, ndataset, channel) TF tensor, if None will just use xyz as

Input: k: int32, number of k SMALLEST elements selected dist: (b,m,n) float32 array, distance matrix, m query points, n dataset p

input: k (1), distance matrix dist (b,m,n) output: idx (b,m,n), val (b,m,n)

input: points (b,m,c), idx (b,n,3), weight (b,n,3) output: out (b,n,c)

input: grad_out (b,n,c), idx (b,n,3), weight (b,n,3) output: grad_points (b,m,c)

Find three nearest neigbors with square distance input: xyz1 (b,n,3), xyz2(b,m,3) output: dist (b,n,3), idx (b,n,3)

Find three nearest neigbors with square distance input: xyz1 (b,n,3), xyz2(b,m,3) output: dist (b,n,3), idx (b,n,3)

Forward pass of a NetVLAD block. Args: reshaped_input: If your input is in that form: 'batch_size' x 'max_samples' x 'feature_size' I