How to plot Grid cell distance from the sensor vs. logarithmically spaced mean number of points per grid cell

[chowkamlee81]

How to plot the graph of

Grid cell distance from the sensor vs. logarithmically spaced mean number of points per grid cell

Kindly help and give pointers

[chowkamlee81]

@YangZhang4065 @edwardzhou130 Kindly help on this. Do the needful

[edwardzhou130]

Hi @chowkamlee81,

You can use grid_ind, which gives you the index of the assigned voxel for each point,

PolarSeg/dataloader/dataset.py

Line 223 in 327fd9b

grid_ind = (np.floor((np.clip(xyz_pol,min_bound,max_bound)-min_bound)/intervals)).astype(np.int)

and voxel_position, which is the position of each voxel.

PolarSeg/dataloader/dataset.py

Line 229 in 327fd9b

voxel_position = np.indices(self.grid_size)*intervals.reshape(dim_array) + min_bound.reshape(dim_array)

Doing the reverse to count the number of points in each voxel and averaging the number of occurrences for voxels in the same distance, you will get the graph.

[torchlidar]

Hi @chowkamlee81,

You can use grid_ind, which gives you the index of the assigned voxel for each point,

PolarSeg/dataloader/dataset.py

Line 223 in 327fd9b

grid_ind = (np.floor((np.clip(xyz_pol,min_bound,max_bound)-min_bound)/intervals)).astype(np.int)

and voxel_position, which is the position of each voxel.

PolarSeg/dataloader/dataset.py

Line 229 in 327fd9b

voxel_position = np.indices(self.grid_size)*intervals.reshape(dim_array) + min_bound.reshape(dim_array)

Doing the reverse to count the number of points in each voxel and averaging the number of occurrences for voxels in the same distance, you will get the graph.

Plot the grid for distance code doesn't open, does it?