Convert predicted box parameters to label format.

Scatter updates into a new (initially zero) tensor according to indices. x (b, feat, k) y index LongTensor (b, k, 3) shape (n1, n2, n

Computes the precision@k for the specified values of k

Collect and dump 3D bounding box statistics

Frustum rotation of 3D bounding box corners.

Frustum rotation of 3D bounding box corners.

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

Get the output directory determined by the current global config.

measure overlap between bird's eye view bounding boxes, parametrized by (ry, l, w, tx, tz)

measure overlap between bird's eye view bounding boxes, parametrized by (ry, l, w, tx, tz)

Log a histogram of the tensor of values.

criterion defines whether the overlap is computed with respect to both areas (ground truth and detection) or with respect to box a or b (detection and

criterion defines whether the overlap is computed with respect to both areas (ground truth and detection) or with respect to box a or b (detection and

Log a list of images.

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

Merge `cfg_other` into the global config.

Select point cloud with predicted 3D mask, translate coordinates to the masked points centroid. Input: point_cloud: (B, 3, N)

C++ interface

Randomly shift box center, randomly scale width and height

Randomly shift box center, randomly scale width and height

a python implementation of SUNRGBDTOOL read3dPoints.m Rtilt: (3, 3) K: (3, 3)

Parse lines in 2D detection output files

dets: [[cx, cy, cz, l, w, h, ry, score]] n,8

dets: [[cx, cz, l, w, ry, score]] n,6

Rotation about the x-axis.

3D Rotation about the x-axis.

Rotation about the y-axis.

Rotation about the z-axis.

Scatter updates into a new (initially zero) tensor according to indices. x (b, feat, k) y index LongTensor (b, k, 3) shape (n1, n2, n

Scatter updates into a new (initially zero) tensor according to indices. x (b, feat, k) y index LongTensor (b, k, 3) shape (n1, n2, n

Project LiDAR points to image

input: B, * target: B, *

image orientation

image orientation

false negatives

false negatives

Transforation matrix from rotation matrix and translation vector.

Transforation matrix from rotation matrix and translation vector.

Write 2D detection results for KITTI evaluation. Convert from Wei's format to KITTI format. Input: det_filename: string, each l

Write 2D detection results for KITTI evaluation. Convert from Wei's format to KITTI format. Input: det_filename: string, each li