Re: competition with outliers and low density point cloud using statisticalOutlierRemovel

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

Re: competition with outliers and low density point cloud using statisticalOutlierRemovel

Wei-2
Hello Everybody,

A question about using the outlier remove in 3D object model building (e.g. milk box) as following picture:
I think <StatisticalOutlierReemoval> can not solve this problem just adjusting k mean and threshold.
Is there any body give me some suggestions to remove this kind of outliers but not influenced the low density part.
Thanks a lot!
Best,
Wei


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

Re: competition with outliers and low density point cloud using statisticalOutlierRemovel

Radu B Rusu
Administrator
Just looking at the data, I would try a MovingLeastSquares (MLS) reconstruction on the data. Check out Alex Ichim's blog
posts on TOCS to learn more about these techniques. (Alex, should we be bringing some of those images into the Doxygen
documentation?).

Cheers,
Radu.

On 06/21/2012 02:54 AM, [hidden email] wrote:

> Hello Everybody,
>
> A question about using the outlier remove in 3D object model building (e.g. milk box) as following picture:
> I think <StatisticalOutlierReemoval> can not solve this problem just adjusting k mean and threshold.
> Is there any body give me some suggestions to remove this kind of outliers but not influenced the low density part.
> Thanks a lot!
> Best,
> Wei
>
>
>
> _______________________________________________
> [hidden email] / http://pointclouds.org
> http://pointclouds.org/mailman/listinfo/pcl-users
_______________________________________________
[hidden email] / http://pointclouds.org
http://pointclouds.org/mailman/listinfo/pcl-users