Point cloud smart mapping in 3D with Scene Viewer

Today, I want to discuss how you can use Scene Viewer to style and configure point clouds directly in your browser, interactively, and intuitively, leveraging the power of smart mapping smart defaults and color ramp sliders. Cities across the world are leveraging the power of point clouds to visualize and present their 3D data, and the smart mapping styles in Scene Viewer give you the ability to customize your visualizations of point cloud data to suite the unique needs of your projects. All you need to do is sign in to ArcGIS Online and you’re on your way.


Explore point cloud styles

To visualize point cloud scene layers, I can choose one of the different smart mapping drawing styles available based on the point cloud attributes: True Color, Class, Elevation, and Intensity. Each style has its own strengths and value depending on the scene I’m trying to create. For example, I can use the Elevation style to explore and understand the changing elevation of the structures and terrain in my scenes.

Hint: The smart mapping styles also include an easy-to-use slider that lets me manually change the point size from smaller to larger and vice versa. Furthermore, if my point cloud layer doesn’t match the terrain, I can apply an offset to my point cloud layer to align it better.


Show your true colors

If I’m interested in displaying the realistic look and feel of my city or landscape, I would choose the True Color style because it displays point cloud data with the true RGB values collected in the LiDAR remote-sensing process. For example, this scene of Mile High Stadium in Denver, Colorado really got my attention. The scene showcases point cloud data created by Merrick and with the True Color style chosen, I can even clearly see the team name “BRONCOS” spelled out in the end zone!


Can I see my point cloud data classified?

With the Class style, I can assign a specific color to the classes in the point cloud data defined in the collection process.  This is helpful when I want to emphasize and distinguish the most important elements of my scene, such as buildings, plants and vegetation, and ground.  When I open the Class style, I can see what classes are available by looking at the list and also in the legend.

Hint: If I turn the “Modulate using intensity “option on, I can shade the color of the points based on the intensity value (see the Intensity style below). This improves depth perception and perspective of my point cloud scene when many nearby points have the same color. By adding this option and playing around with the class colors, I can see much more detail, such as the roads and roof structures.

Also, in case you’re wondering what those trees are doing on the highway, please don’t call the forest service. They’re actually cars. However, LiDAR scanning is not a perfect science and sometimes automated classification fails and puts some points in the wrong category.


What I’d really like to see is the elevation of my points

I chose the Elevation style to visualize this Helsinki scene I created to emphasize the higher areas and structures. Knowing the elevation of buildings and terrain throughout Helsinki is very important from an urban planning, environmental, or industrial perspective. For example, if I am in the shipping or utilities industries, I may want to focus on structures higher than a certain height, such as cargo cranes or power lines. I can adjust the Elevation style sliders to make the higher areas and structures stand out more.


What is intensity in point clouds?

Using the Intensity style is another great way to see the fine details and textures of Helsinki’s urban landscape. Intensity is a measure of the return strength of the laser pulse that generated the points in the point cloud data collection process. With the Intensity style on, I can get even more detail and contrast in my scene by adjusting the slider to match the distribution of point values.

First, I drop down the top slider bar to 190 intensity value. Then I set the maximum value to 200 to reset the distribution of the color ramp.

Now I can really see the contrast between the buildings, streets, trees, and canals. All that’s left for me to do is pick out a pleasing color ramp and my scene is finalized.

I encourage you to explore for yourself the power of smart mapping with point clouds in Scene Viewer. To learn more about point cloud smart mapping styles, see style point clouds. And feel free to share your own cool scene creations and comments.

Scenes and screenshots:

This entry was posted in 3D GIS, ArcGIS Online, Local Government, Web and tagged , , , , , . Bookmark the permalink.

Leave a Reply


  1. lisaedman says:

    When will this be availiable in portal for arcgis?

    • Matt Miller says:

      Thanks for your comment. ArGIS Enterprise 10.6 (releasing around January) will have smart mapping with point clouds in Scene Viewer plus a whole lot more features, such as building scene layer smart mapping and annotating. Stay tuned.

  2. lebeno777 says:

    thanks,esri good luck!!!!

  3. dougbrowning says:

    How do you suggest getting 600+GB of las point files up to AGO? Thanks

    • Matt Miller says:

      That is something we’d be happy to help you out with. Let me talk to a few folks here and get back to you.

    • The process of getting 600+GB of LAS data on ArcGIS Online will require the creation of a LAS Dataset in Pro. Once the LAS dataset is created using the “create scene layer package” geoprocessing tool you can create a Scene Layer Package (SLPK). With the SLPK created using the “share package” Geoprocessing tool you can share and publish the data to your ArcGIS Online organization account.

      • dougbrowning says:

        Why does Pro not let you map to a root drive? My dataset is on a usb drive at H:\dataset.lasd. Pro has no way to map to the root of H: so I cannot use my dataset. If I move it all the paths are wrong. Weird.

        Also note you must add the dataset to a map then add to the tool. If you try to pick directly it will not take it.

        • dougbrowning says:

          After 7.5 hours I got this error. Out files were of 0 size. Any Ideas? Thanks

          [Info]: Pass1 file hfm_02749.las [2808/2808]
          [Info]: Pass2 file hfm_00020.las [1/2808]
          [Critical]: Failed to compress data using method [2]
          Failed to Execute
          Failed to execute (CreateSceneLayerPackage).

          • When you say 0 size do you mean there was no data in the files?
            Remove the listed tiles using the LASD properties console and run again.
            To quickly verify create a small LASD with the offending tiles and run.

        • You may add a folder connection in the catalog tab in Pro which maps that drive to Pro. All paths in you LASD dataset should use relative paths to avoid loosing path linkage.

          The LASD dataset must be added to the map to insure it honors rendering and symbolization. Here is a python script which permits automation of SLPK creation. http://pro.arcgis.com/en/pro-app/tool-reference/data-management/create-scene-layer-package.htm

          • dougbrowning says:

            I did try a smaller dataset and it is still dying on Pass 2 every time. The output .slpk files are always of 0 size when it dies. Sorry i wish I could post screen shots here.

            [Info]: Pass1 file hfm_02748.las [1510/1511]
            [Info]: Pass1 file hfm_02749.las [1511/1511]
            [Info]: Pass2 file hfm_01326.las [1/1511]
            [Info]: Pass2 file hfm_01345.las [2/1511]
            [Critical]: Failed to compress data using method [2]
            Failed to Execute
            Failed to execute (CreateSceneLayerPackage).

          • dougbrowning says:

            Turned out to be the Lidar contractor used a NAD83 (2011) version of UTM instead of the regular UTM. Arc read this as a custom projection. That then crashed the tool.

            I was able to use the 3D tool set and the Create PRJ for LAS tool to set it to regular UTM. Then it all worked.

            Hope it helps someone.