Compute objectness loss for the proposals.

Dump point clouds Args: point_clouds (_type_): _description_ dump_dir (str, optional): _description_. Defaults to "./dump/".

Dump results. Args: end_points: dict {..., pred_mask} pred_mask is a binary mask array of size (batch_size, num_

pc: (N,3), box3d: (8,3)

Flip X-right,Y-forward,Z-up to X-right,Y-down,Z-forward Input and output are both (N,3) array Tensor version.

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

Calculate the Intersection over Union (IoU) of two 2D bounding boxes. Parameters ---------- bb1 : dict Keys: {'x1', 'x2', 'y

Args: error: Torch tensor (d1,d2,...,dk) Returns: loss: Torch tensor (d1,d2,...,dk) x = error = pred - gt or dist(pred,g

Log a list of images.

initiate randomness. :param config: :return:

initiate randomness. :param config: :return:

Parse groundtruth labels to OBB parameters. Args: end_points: dict {center_label, heading_class_label, heading_residual_

input is Nx3 points output is imgsize*imgsize*num_sample*3 assumes points are in range [-radius, radius] samples num_sample p

input is Nx3 points. output is vsize*vsize*vsize assumes points are in range [-radius, radius]

input is Nx3 points output is vsize*vsize*vsize*num_sample*3 assumes points are in range [-radius, radius] samples num_sample

points is a Nx3 numpy array

if string s represents an int.

Rotation about the y-axis.

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

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

Log a scalar variable.

Numpy function for sigmoid

PyTorch Implementation for tf.nn.sigmoid_cross_entropy_with_logits: max(x, 0) - x * z + log(1 + exp(-abs(x))) in https://www.

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

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

Numpy function for softmax

Numpy function for softmax

ap = voc_ap(rec, prec, [use_07_metric]) Compute VOC AP given precision and recall. If use_07_metric is true, uses the VOC 07 11 point met

vol is occupancy grid (value = 0 or 1) of size vsize*vsize*vsize return Nx3 numpy array.

Create lines represented as cylinders connecting pairs of 3D points Args: pcl: (N x 2 x 3 numpy array): N pairs of xyz pos