Segmentation fault using VFH class

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

Segmentation fault using VFH class

Frits Florentinus
Hi,

When I try to compute the VFH of a large point cloud I get a segmentation fault.

442617 points does not fail
4426172 points does

I am pretty sure I implemented the class correctly since the 400k version gives proper results. I am also thinking it is not due to my memory failing: I am working with other PCL classes on even larger data sets and they don't fail.

Further info: I am using pcl::PointXYZ, pcl::Normal and pcl::VFHSignature308 as templating for the VFH class and use a KdTree searcher. So basically it is a direct implementation of the tutorial: http://pointclouds.org/documentation/tutorials/vfh_estimation.php#vfh-estimation

I am not 100% if this is a bug or some issue on my end. Any help would be appreciated.

Frits
Reply | Threaded
Open this post in threaded view
|

Re: Segmentation fault using VFH class

Radu B Rusu
Administrator
Frits,

What's the difference between the two clouds? Does one of them contain NaN data?

Cheers,
Radu.
--
http://pointclouds.org

On 03/29/2012 07:31 AM, Frits Florentinus wrote:

> Hi,
>
> When I try to compute the VFH of a large point cloud I get a segmentation
> fault.
>
> 442617 points does not fail
> 4426172 points does
>
> I am pretty sure I implemented the class correctly since the 400k version
> gives proper results. I am also thinking it is not due to my memory failing:
> I am working with other PCL classes on even larger data sets and they don't
> fail.
>
> Further info: I am using pcl::PointXYZ, pcl::Normal and pcl::VFHSignature308
> as templating for the VFH class and use a KdTree searcher. So basically it
> is a direct implementation of the tutorial:
> http://pointclouds.org/documentation/tutorials/vfh_estimation.php#vfh-estimation
>
> I am not 100% if this is a bug or some issue on my end. Any help would be
> appreciated.
>
> Frits
>
> --
> View this message in context: http://www.pcl-users.org/Segmentation-fault-using-VFH-class-tp3867843p3867843.html
> Sent from the Point Cloud Library (PCL) Users mailing list archive at Nabble.com.
> _______________________________________________
> [hidden email] / http://pointclouds.org
> http://pointclouds.org/mailman/listinfo/pcl-users
_______________________________________________
[hidden email] / http://pointclouds.org
http://pointclouds.org/mailman/listinfo/pcl-users
Reply | Threaded
Open this post in threaded view
|

Re: Segmentation fault using VFH class

Frits Florentinus
No,

the smaller is a downsampled version by a factor 10 of the original. They both do not contain any NaNs.
Reply | Threaded
Open this post in threaded view
|

Re: Segmentation fault using VFH class

Radu B Rusu
Administrator
Frits,

Can you send me the dataset offline? I'll try to replicate this and fix it if it's broken. Ping me on Skype.

Cheers,
Radu.

On 03/29/2012 10:02 AM, Frits Florentinus wrote:

> No,
>
> the smaller is a downsampled version by a factor 10 of the original. They
> both do not contain any NaNs.
>
> --
> View this message in context: http://www.pcl-users.org/Segmentation-fault-using-VFH-class-tp3867843p3868337.html
> Sent from the Point Cloud Library (PCL) Users mailing list archive at Nabble.com.
> _______________________________________________
> [hidden email] / http://pointclouds.org
> http://pointclouds.org/mailman/listinfo/pcl-users
_______________________________________________
[hidden email] / http://pointclouds.org
http://pointclouds.org/mailman/listinfo/pcl-users
Reply | Threaded
Open this post in threaded view
|

Re: Segmentation fault using VFH class

pranny
Any update on this? I am getting the same error (Segmentation Fault) while trying to compile the test code at http://pointclouds.org/documentation/tutorials/vfh_estimation.php.