represent 8 corner points as box parameters (center, size, heading_angle) 6 -------- 5 /| /| 7 -------- 4 .

Cosine rampdown from https://arxiv.org/abs/1608.03983

Draw 3D bounding boxes Args: gt_boxes3d: numpy array (n,8,3) for XYZs of the box corners fig: mayavi figure handler color

Draw 3d bounding box in image qs: (8,2) array of vertices for the 3d box in following order: 1 -------- 0 /| /

Generic functions to compute precision/recall for object detection for multiple classes. Input: pred_all: map of {img_id:

r""" Uses iterative furthest point sampling to select a set of npoint features that have the largest minimum distance Paramet

r""" Parameters ---------- features : torch.Tensor (B, C, N) tensor idx : torch.Tensor (B, n

r""" Find the three nearest neighbors of unknown in known Parameters ---------- unknown : torch.Tensor

r""" Performs weight linear interpolation on 3 features Parameters ---------- features : torch.Tensor

r""" Parameters ---------- features : torch.Tensor (B, C, N) tensor of features to group idx : torch.Tens

r""" Parameters ---------- radius : float radius of the balls nsample : int maximum number of

r""" Parameters ---------- xyz : torch.Tensor xyz coordinates of the features (B, N, 3) new_xyz : torch.Te

r""" Parameters ---------- xyz : torch.Tensor xyz coordinates of the features (B, N, 3) new_xyz : torch.Te

r""" Parameters ---------- xyz : torch.Tensor (B, N, 3) tensor of the xyz coordinates of the features feat

r""" Parameters ---------- xyz : torch.Tensor (B, N, 3) tensor of the xyz coordinates of the features feat

r""" Parameters ---------- xyz : torch.Tensor (B, N, 3) tensor of the xyz coordinates of the features feat

r""" Parameters ---------- unknown : torch.Tensor (B, n, 3) tensor of the xyz positions of the unknown features

r""" Propagate features from xyz1 to xyz2. Parameters ---------- xyz2 : torch.Tensor (B, N2, 3) tensor of the xyz

r""" Forward pass of the network Parameters ---------- pointcloud: Variable(torch.cuda.FloatTensor)

Forward pass of the network Args: inputs: dict {point_clouds} point_clouds: Variable(torch.cuda

Forward pass. Arguments: seed_xyz: (batch_size, num_seed, 3) Pytorch tensor seed_features: (batch_size, feature_dim,

Args: xyz: (B,K,3) features: (B,C,K) Returns: scores: (B,num_proposal,2+3+NH*2+NS*4)

box_size: [x1,x2,...,xn,3] heading_angle: [x1,x2,...,xn] center: [x1,x2,...,xn,3] Return: [x1,x3,...,xn,8,3]

box_size is array(l,w,h), heading_angle is radius clockwise from pos x axis, center is xyz of box center output (8,3) array for 3D box cornde

Args: end_points: dict { center, size_scores, size_residuals_normalized, sem_cls_scores, flip

Loss functions for supervised samples in training detector Args: end_points: dict { seed_xyz, seed_inds, vot

Compute IoU of two bounding boxes. ** Define your bod IoU function HERE **

Generate ROI point cloud w.r.t predicted box :param inputs: dict {'point_clouds'} input point clouds of the whole scene

Compute number of objects for each semantic class

Log a histogram of the tensor of values.

Inverse a rigid body transform matrix (3x4 as [R|t]) [R'|-R't; 0|1]

Linear rampup

Args: labels: (B, N, 6) pred: (B, M, 6) iou_thresh: scalar conf_thresh: scalar label_mask: (B, N,) with v

Input is BxNx3 a batch of point cloud Output is BxIxIxnum_samplex3 Added on Feb 19

Input is BxNx3 batch of point cloud Output is Bx(vsize^3)

Input is BxNx3 a batch of point cloud Output is BxVxVxVxnum_samplex3 Added on Feb 19

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

Randomly shift box center, randomly scale width and height

read XYZ for each vertex.

read XYZ RGB for each vertex. Note: RGB values are in 0-255

read XYZ point cloud from filename PLY file

Input ps is NxC points with first 3 channels as XYZ z is facing forward, x is left ward, y is downward

Input: (n,3), Output: (n,3)

Rotation about the x-axis.

Rotation about the y-axis.

Rotation about the y-axis.

Rotation about the y-axis. t: (x1,x2,...xn) return: (x1,x2,...,xn,3,3)

Rotation about the z-axis.

Exponential rampup from https://arxiv.org/abs/1610.02242

Convert 3D box size (l,w,h) to size class and size residual

Transforation matrix from rotation matrix and translation vector.

Export scene bbox to meshes Args: scene_bbox: (N x 6 numpy array): xyz pos of center and 3 lengths out_filename: (string) filename

Export oriented (around Y axis) scene bbox to meshes Args: scene_bbox: (N x 7 numpy array): xyz pos of center and 3 lengths (dx,dy,dz)

input: Nx3, write points to filename as PLY format.

Color (N,3) points with labels (N) within range 0 ~ num_classes-1 as OBJ file

Color (N,3) points with RGB colors (N,3) within range [0,255] as OBJ file