'problems with computing normal for organized point cloud

Now I want using PCA method to compute normal for the organized point cloud transformed from the depth image.Here is what I do:

pcl::NormalEstimation<pcl::PointXYZ, pcl::Normal> normObj;

normObj.setInputCloud (cloud);  //cloud is an organized point cloud  

pcl::search::OrganizedNeighbor<pcl::PointXYZ>::Ptr tree(new pcl::search::OrganizedNeighbor<pcl::PointXYZ>());  

normObj.setSearchMethod (tree);

//normObj.setRadiusSearch (0.05); 

normObj.setKSearch(10);   

PointCloud<pcl::Normal>::Ptr normals (new PointCloud<pcl::Normal>); 

normObj.compute (*normals); 

the problem is when I run the program, there will be no result but without giving an error.What's wrong with the code?



Solution 1:[1]

This might be too late but for people who are struggling this in the future, the structured point cloud normal can be estimated using IntegralImageNormalEstimation I am not sure the normal quality but it works for me.

Sources

This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.

Source: Stack Overflow

Solution Source
Solution 1 Zhaozhong Chen