Does transformation invalidate the computed features

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Does transformation invalidate the computed features

Humam
Hi,
If the features(FPFHSignature33) of a point cloud was computed. Then, the
point cloud was transformed using pcl::TransformPointCloud.

Does this invalidate the computed features? Should I re-compute them again
for the transformed point cloud?
Thanks



--
Sent from: http://www.pcl-users.org/
_______________________________________________
[hidden email] / http://pointclouds.org
http://pointclouds.org/mailman/listinfo/pcl-users
Reply | Threaded
Open this post in threaded view
|

Re: Does transformation invalidate the computed features

MarkL
FPFH is probably to *some degree* transformation invariant. I don't remember
the technical details of it's implementation. You may read in the tutorial
and the corresponding article.
http://pointclouds.org/documentation/tutorials/fpfh_estimation.php

However, I can tell you from my experience of using it for alignment in my
particular data, that doing a prior transformation of the input cloud (very
coarse alignment by normals) had a very positive effect on the performance
of the following step (alignment using FPFH).

I wouldn't go as far as to say it invalidates the features, but I think
rec-computing the features can help to finetune them.


Humam wrote
> If the features(FPFHSignature33) of a point cloud was computed. Then, the
> point cloud was transformed using pcl::TransformPointCloud.
>
> Does this invalidate the computed features? Should I re-compute them again
> for the transformed point cloud?
> Thanks





--
Sent from: http://www.pcl-users.org/
_______________________________________________
[hidden email] / http://pointclouds.org
http://pointclouds.org/mailman/listinfo/pcl-users