www.my3dscanner.com

Accurate Dense 3D Point Clouds in Full Color

Accurate Dense 3D Point Clouds in Full Color

The quality of 3D scans made with your camera could be compatible with expensive 3D scanners which cost thousands of dollars. And only few of them can scan in color. MY3DSCANNER certainly can! And how much does it cost? It’s free! You probably already have a professional 3D scanner in your pocket!

Sophisticated Technology Behind Simple and Friendly Interface

Sophisticated Technology Behind Simple and Friendly Interface

Rapid advances in computer vision technologies stand behind MY3DSCANNER computational engine. We just made it user-friendly for you. The process is fully automatic: simply take and upload your photos to this site.  MY3DSCANNER will do the rest!

Become 3D Artist with My3Dscanner

Become 3D Artist with My3Dscanner

Turn your photos into beautiful 3D models. Create rich 3D content for visualization. Make 3D while you are travelling.  You do not need to be a professional 3D modeler:  instead of expensive software use MY3DSCANNER to create realistic 3D model of virtually anything. Applications in virtual world or for manufacturing are endless!

Use 3D Models for Business

Use 3D Models for Business

3D scanning creates blue-prints for manufacturing.  Use point cloud directly for 3D laser sub-surface engraving.  Turn you 3D model into CNC-machined product or a mold. Use your 3D models for rapid prototyping.  Make physical replica of the world! This is the place to start or improve your business with no investments at all!

Your Camera or Mobile Phone is Already a 3D Scanner

Your Camera or Mobile Phone is Already a 3D Scanner

It is as incredible as it sounds. Take photographs with digital camera or mobile phone and receive dense point cloud representing the object in 3D. MY3DSCANNER is on-line service to create 3D scans for free and make 3D models from point clouds.

Its's Easy to Start. It's Free. Join the Community!

Its's Easy to Start. It's Free. Join the Community!

Read quick guide, take photos and upload them to MY3DSCANNER. Receive 3D dense point cloud and turn it into complete 3D model if you want.  Show your projects to people. Share your 3D scanning experience. Get advice or comment other projects.

Frontpage Slideshow | Copyright © 2006-2010 JoomlaWorks, a business unit of Nuevvo Webware Ltd.
Monday, 21 February 2011 11:09

How to Integrate Point Cloud into Google SketchUp

Written by 
Rate this item
(8 votes)

You can create an accurate blueprint for modeling of a real object in SketchUp using free 3D photo-based scanner. Points created with www.my3dscanner.com  will serve as construction marks for SketchUp model.  Later you can link construction points automatically or manually to create complex polygonal replicas of the real world in Sketch Up.

Let me guide you how to do it. There are 4 main steps:

1.       Scan an object and receive a 3D point cloud (if you don’t have one yet)

2.       Reduce the density of the point cloud (optional but recommended)

3.       Turn the point cloud into Sketch Up readable format

4.       Import the point cloud into Sketch Up

 You will need the following free (except MS Excel but use their trial or borrow) tools:

www.my3dscanner.com to create 3D scan of a real object with your digital camera. It is free on-line scanning service (no need to download)

MeshLab is free software to view and manage point clouds (optional) - download

Microsoft Excel

Google SketchUp

Cloud_v71.rb  is SU7 plug-in - download

Workflow:

Step 1: Scan an object - create point cloud

Create a 3d project (e.g. my3dproject) with your digital camera or a mobile phone at www.my3dscanner.com :

  • Read tutorials
  • Register
  • Take and upload photos
  • Download  result: a zipped 3D point cloud in  .ply format. Download my3dprogect.zip   to your computer from My Studio/My Projects.
  • Unzip it to have your point cloud my3dproject.ply file ready for SketchUp integration.

Step 2: Reduce the density of your point cloud (optional but recommended step)

Your point cloud most likely will be very dense. It is good for visualization but bad for Sketch Up modeling. You need to end up with 2-5,000 points maximum. To sub-sample a point cloud to make it less dense use MeshLab:

File -> Open -> my3dproject.ply

Filters -> Sampling -> Clustered vertex Subsampling  -> Dialog Box.

Dialog Box: Apply, Close

This way you will apply Default Values in a this dialog box. If you will to sub-sample only some parts of your point cloud, check “Select” box. You can play with values in the dialog boxes (all your attempts will be saved in different layers as described below). Experiment!

Apparently you will not see the result immediately.

To view it, go to View -> Show Layer Dialog.

Switch visibility off for my3dproject in the right window. You will be left with Cluster Samples (in yellow) layer which is less dense point cloud. This is exactly what we wanted to achieve.

Click on Cluster Samples and:  File -> Save As - > Name: my3dproject, format: Stanford Polygon Format (*ply) -> Save

In the next pop-up window unmark Binary encoding and click OK.

Now you have your my3dproject.ply point cloud reduced.

Step 3: Turn your point cloud into csv file

Start Excel: File -> Open -> my3dproject.ply  -> Text Import Wizard:

 

Text Import Wizard Step 1

 

Check Delimited

 

Change “Start from the line” to 11 (if you have skipped step 2 of this tutorial, change to 9). You are just deleting the header of the file, so adjust the row number respectively.

 

Do not change other settings. Next

 

Text Import Wizard Step 2

 

Check Space -> Next

Text Import Wizard Step 3

Check General -> Finish

You will have a table with 3 rows of digits (or more if you have skipped Step 2)

We are almost done:

File -> Save As -> Other Formats -> File name: my3dproject, File type: СSV (Comma delimited) (*.csv) -> Save -> Yes

Now you have my3dproject.csv ready.

 

Step 4: Import .csv  into SketchUp

Download and install cloud_71.rb plugin into Google SketchUp. It will add an option Points Cloud in File menu of SketchUp.

 

Start Sketch Up and import point cloud:

 

File -> Points Cloud -> Import - > my3dproject.csv -> OK -> Dialog Box “Points Cloud Options”

 

Data delimiter:                 ;

Import unit:                    Meters

Flatten Z coordinate:        No

 

 

Click OK. Make sure the values you enter are the same as in the screenshot above.

 

Click OK in the next screen.

 

Then click OK in “Additional Options” window.

 

Wait, do not rush Sketch Up!

 

The last step is a window “Triangulate the point cloud?”

 I’d rather say NO since there are better options to triangulate (I will talk about them  in my next tutorials, ) but you can also click “yes” and wait (SketchUp may crash at this point of time if point cloud is too big). If you do not like the triangulation/mesh result, just kill the layer, not the construction points.

This is it. Your scan is in SketchUp. It is not as pretty as in www.my3dscanner.com (you will get sparse point cloud with no color), but real builders use crude concrete and steel bars, not artistic colored points, right?

Do you want to triangulate/mesh the point cloud in SketchUp?  The next tutorial is coming!

6 comments

  • Comment Link Paul Wednesday, 18 September 2013 12:02 posted by Paul

    Dustin, it's a rather complex and lengthy method. I really can't tell you what had gone wrong given your sparse description of the problems you had faced.

  • Comment Link Dustin Tuesday, 25 June 2013 20:37 posted by Dustin

    I've gone though all of the steps yet when I import the csv file into sketch up it does not show the sample line of data in the opening window.

    I've set up my .csv with tabs and spaces and everything seams to be working.

    Is this because I have to many points in my model? Where am I going wrong?

    Thanks in advance

  • Comment Link Dustin Tuesday, 25 June 2013 20:34 posted by Dustin

    enter your message here...

  • Comment Link TK Thursday, 09 February 2012 20:56 posted by TK

    I appreciate your help

  • Comment Link Jayan Saturday, 23 April 2011 03:58 posted by Jayan

    That's the best ansewr of all time! JMHO

  • Comment Link Christina Friday, 22 April 2011 17:41 posted by Christina

    That's way more celevr than I was expecting. Thanks!

Login to post comments