1

What are the limits of viewing points and rasters together in QGIS 3D (either in-app or using qgis2threejs) in comparison to ArcScene?

Is using GRASS NVIZ the solution? (see end of post)

I am using QGIS 3.4.3. I am an archaeologist and I would like to appropriately 3D view the Z relationships between (1) total station-derived points (with XYZ attributes), which I took on multiple superimposed excavation layers, along with (2) DEM or TIN rasters. These ts vector points and DEM/TIN rasters should be superimposed like a layer cake, because they are excavated surfaces that were documented sequentially as dirt was removed.

When I tried qgis2threejs (2.3.1) the DEMs/TINs are anchored way below my points when I use any vertical exaggeration (for example, .5), which is needed to see my points in three dimensions. If I don't use any vertical exaggeration, everything is flat. I'm assuming the DEM is anchored at 0 masl, whereas my points are over 3200 masl (z values). (Note that I followed the plugin's documentation and converted all file and project CRMs to 3857.) See Image:

Towered DEM with point shapefile far above

I simply want to see the DEM displayed using its cell (Z) values so I can see its true relationship to the total station points, such as is possible in ArcScene. Here is an example of this in ArcScene (not the best example, but it shows how there is a DEM floating between a cluster of points, at its appropriate height):

ArcScene DEM and point shapefile

When I use the built in QGIS 3D viewer, it seems I am only able to view a raster clamped to a generic place in space. My points will drape over the raster, but I need to see them in their appropriate XYZ location. If choose "Absolute" for the Altitude Clamping, the points are then moved below the DEM raster (while also shown draped on the DEM surface...):

QGIS built-in 3D viewer, with points shown below DEM (note that an impression of the points are also shown glued to the DEM surface, but that's not showing the points' location relative to the DEM surface)

I am confused by this, because the 3D viewer (in settings) wants/requires you to chose one of your available rasters for the 3d image, but then it doesn't seem to recognize its Z position. For points, however, it allows you to display them using their "absolute" altitude (Z) in the layer properties, but these points can't be shown in relation to the DEM because the DEM is not visualized at its "absolute" altitude. (note: in settings, I tried selecting no raster (the blank line at the top of the drop-down list of rasters) and when I do this, the points go way above the DEM (rather than below as before), but the model is useless to navigate--everything is tiny or disappears when you zoom, and the view jumps everywhere when you try to turn the model.)

Can neither QGIS 3D view nor the qgis2threejs plugin show rasters and vector shapefiles together in their absolute positions like ArcScene can?

If so, it is good to know that this is a limitation of QGIS and I should stop looking for ways around it (I respect that the 3D viewer is newish).

Or is it that I am just not understanding how to appropriately adjust settings for the rasters?

If it is a limitation of QGIS, are there any open source or affordable software options that I can use to easily view the superposition of multiple points and raster layers together in 3D?

Bonus would be to also be able to click on vector points to identify their attributes: most important for my need is to be able to review and explore the data rather than to create a publishable image. I've migrated nearly completely to QGIS from ArcGIS, but I may have reached a wall with this.


Having done further research to find a solution/alternative, it seems that I should be able to use NVIZ to visualize in 3D within QGIS to visualize several layers of different kinds at once, and NVIZ appears to be more sophisticated than QGIS's 3D view or qgis2threejs. But I've not yet mastered how to use it: the log after running NVIZ ends with "Loading resulting layers; "Algorithm 'nviz' finished" but no window pops up to show the 3D data as the process should (apparently) do.

closed as too broad by PolyGeo Mar 21 at 22:03

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • As per the Tour and Help center there should be only one question asked per question and questions should not request opinions as answers. – PolyGeo Mar 21 at 22:05
0

**You could try setting the properties of your point layer. **

  1. On the Qgis2threejs window exporter, right click on your point layer
  2. On the Layer Properties window, select "Z Value" and then, the field that contains the Z Valeus. Click Apply

enter image description here

enter image description here

For a better visualization, you can use Blender, and Blender GIS.

See Contour lines with imported shapefile using Blender GIS - wrong Z values

enter image description here

  • None of the blender examples show data under a surface, is it possible? – nmtoken Mar 20 at 20:05
  • @nmtoken yes, sure. It is just a matter of importing different SHPs into Blender. Blender is a complete 3D modelling software, so you can do anything related to 3D with it. – Marcos Saito Mar 21 at 11:26
  • Thank you for the help @Marcos. I had, in fact, changed the properties of the point vector layers to show elevation using its absolute Z value -- that's precisely my issue: that the points, with their absolute Z-values used, aren't shown relative to the cell velues of the raster. I've also played around with Blender GIS and it seems that my data may be too extensive for it to be useful (I've read that scale is a limitation with Blender GIS's capabilities) and I also don't think I can access the vector attributes in Blender (but I do find using Blender helpful for other uses, thanks). – ArchArchy Mar 21 at 17:15
  • How many points are you talking about in your SHP? And what size is the raster? Try making a clipping at both your points SHP and the raster (begin with just 2 or 3 points to see if there is something wrong, then scale up and see if a pattern emerges). It is always easier to understand what is going on with just a few things. If you want, provide a link with a sample of the SHP and the raster, so I can try it on my computer. – Marcos Saito Mar 22 at 18:08

Not the answer you're looking for? Browse other questions tagged or ask your own question.