Fast 3D point cloud segmentation using supervoxels with geometry and color for 3D scene understanding
Home Page: https://ieeexplore.ieee.org/abstract/document/8019382
License: BSD 3-Clause "New" or "Revised" License
Thanks for making your work available.
Is there any way to save the segmented/colored point cloud?
When I try to make it, I get this error:
supervoxel_clustering/src/testing.cpp:157:66: error: conversion from ‘boost::detail::sp_if_not_array<pcl::PointCloud<pcl::PointXYZL> >::type {aka boost::shared_ptr<pcl::PointCloud<pcl::PointXYZL> >}’ to non-scalar type ‘pcl::PointCloud<pcl::PointXYZL>::Ptr {aka std::shared_ptr<pcl::PointCloud<pcl::PointXYZL> >}’ requested
PointLCloudT::Ptr subcloud = boost::make_shared<PointLCloudT>();
Since this was in the testing.cpp, I tried ignoring this file to build the source. But it died again on:
supervoxel_clustering/src/clustering.cpp:406:63: error: conversion from ‘boost::detail::sp_if_not_array<pcl::Supervoxel<pcl::PointXYZRGBA> >::type {aka boost::shared_ptr<pcl::Supervoxel<pcl::PointXYZRGBA> >}’ to non-scalar type ‘pcl::Supervoxel<pcl::PointXYZRGBA>::Ptr {aka std::shared_ptr<pcl::Supervoxel<pcl::PointXYZRGBA> >}’ requested
SupervoxelT::Ptr sup_new = boost::make_shared<SupervoxelT>();
What could be going wrong with the boost::make_shared instantiation?
Boost Version: 1.58.0.1ubuntu1
PCL Version: 1.11
C++14 compiler
I compile successfully and run the given pcd demo:
./supervoxel_clustering -p ../pcd/milk_cartoon_all_small_clorox.pcd
Here is the segmentation visualization result:
It doesn't look great.
And this is my terminal output information:
Loading pointcloud from PCD file '../pcd/milk_cartoon_all_small_clorox.pcd'...
Failed to find match for field 'label'.
Pointcloud loaded
Extracting supervoxels...
Found 604 supervoxels
Getting supervoxel adjacency...
Refining supervoxels...
Constructing Boost Graph Library Adjacency List...
Segmentation initialization...
Testing thresholds from 0.800000 to 1.000000 (step 0.005000)
<T, Fscore, voi, wov> = <0.800000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.805000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.810000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.815000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.820000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.825000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.830000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.835000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.840000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.845000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.850000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.855000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.860000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.865000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.870000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.875000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.880000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.885000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.890000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.895000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.900000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.905000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.910000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.915000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.920000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.925000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.930000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.935000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.940000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.945000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.950000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.955000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.960000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.965000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.970000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.975000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.980000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.985000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.990000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <0.995000, 0.996362, 0.047721, 0.992751>
<T, Fscore, voi, wov> = <1.000000, 0.996362, 0.047721, 0.992751>
Using best threshold: 0.800000 (F-score 0.996362, voi 0.047721)
Initialization complete
Starting clustering...
Clustering complete
Initializing testing suite...
Loading visualization...
Loading viewer...
Scores:
VOI 0.047721
Prec. 1.000000
Recall 0.992751
F-score 0.996362
WOv 0.992751
FPR 0.000000
FNR 0.007249Is there any problem? Would you like to share your output details?
Thanks!
When I use your pcd file in your files without ROS.
In my terminal I input "build/supervoxel_clusrering -p pcd/milk_cartoon_all_small_colorx_pcd", ther is a tip in my terminal "Failed to find match for field 'label'. And the other step is good.
Last, my visulize window is only has the supervoxel normals result(whatever I choose 0-4 is ON or only choose 3 or others ), and no have others result. No color, only white normal vector.
What's wrong? Hope your answer.
excuse me, I have run your code many times, but the segmentation effect is not ideal, because there are too many parameters, so I would like to ask what are the specific parameters of the runtime in your example: milk_cartoon_all_small_clorox.pcd.
Hope to get your reply, thank you!
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
Django
The Web framework for perfectionists with deadlines.
Laravel
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft
Open source projects and samples from Microsoft.
Google
Google ❤️ Open Source for everyone.
Alibaba
Alibaba Open Source for everyone
D3
Data-Driven Documents codes.
Tencent
China tencent open source team.