shiaoming / stair-perception

A stair detection and modeling implementation using point cloud data with PCL(Point Cloud Library).

Hi there 👋

stair-perception's Issues

Error with own point cloud

Hello, good job!

We are currently using the stair-perception to analyze our own point cloud, but we get the following error:

Can you help us here?

Thank you!

A little mistake I found ?

@Shiaoming

I think I found a little mistake in the function StairDetection::computeVectorPlaneInfo

stair-perception/modules/stairperception/StairPerception.cpp

Lines 1313 to 1316 in b498955

	// 4.middle point of principle vector
	plane.pcenter.x = (points_max[0].x + points_max[0].x)/2;
	plane.pcenter.y = (points_max[0].y + points_max[0].y)/2;
	plane.pcenter.z = (points_max[0].z + points_max[0].z)/2;

I guess you want to calculate every plane's principle centre after we get the points_max and points_min. And I think what you want to do is as follow.

plane.pcenter.x = (points_max[0].x + points_min[0].x)/2;
plane.pcenter.y = (points_max[0].y + points_min[0].y)/2;
plane.pcenter.z = (points_max[0].z + points_min[0].z)/2;

I am not sure if I am right or not ?

求教

博主，你好．可以发下那篇文章吗？求帮助．会议论文在网上没刊登

does it work for downstairs ?

hi,
thanks for your work.
Just wondering if this package is also applicable for downstairs. what are your thoughts about it?

How to detect the descent stair

Hello, This is a wonderful job. But I confused how to detect the descent stair

A little doubt about the test time

Hi authors,

This stairs-perception project is really interesting and amazing.
My Ubuntu 16.04 equips with an i5-8600K 3.6GHz CPU and I tested the project with the pcd file provided in the "sample" folder.

As you can see from the picture, the total consuming time was 200ms around, which is quite different from the test time given in your paper.

I am really confused about that because the CPU on my computer seems better than the CPU used in the paper but achieved slower results. I wonder if you used GPU to achieve that speed or some other acceleration techniques?

Thanks anyway.

Two small questions on computing line from 2 planes

Hi @Shiaoming
Sorry to bother you again.
This time I ran into a small question when reading the source code.
In the function

stair-perception/modules/stairperception/StairPerception.cpp

Lines 1123 to 1127 in 3cd6222

	// line : (x-x0)/n1=(y-y0)/n2=(z-z0)/n3=t
	// plane: x0*n1+y0*n2+z0*n3=0
	// compute cross point between line and plane perpendicular to this line with in origin
	float t = -1*(normal_line[0]*x0+normal_line[1]*y0+normal_line[2]*z0)
	/ (normal_line[0]*normal_line[0]+normal_line[1]*normal_line[1]+normal_line[2]*normal_line[2]);

you want to compute the intersection line of 2 planes. I am not quite understand why you calculate parameter t in this way.
I know t is the parameter in the vector equation of line, and I just wonder the reason why t is calculated in this way.

Then t is used to calculate another point on the line, whose coordinate is represented by variable x y z .

stair-perception/modules/stairperception/StairPerception.cpp

Lines 1129 to 1131 in 3cd6222

	float x = normal_line[0]*t+x0;
	float y = normal_line[1]*t+y0;
	float z = normal_line[2]*t+z0;

And another place that puzzles me is that what is the meaning of line.h and line.d ?

stair-perception/modules/stairperception/StairPerception.cpp

Lines 1140 to 1141 in 3cd6222

	line.h = x;
	line.d = sqrt(y*y+z*z);

Does it refer to the height and width of the steps , just like the annotation I made in the following picture ?

Thanks !