Aligning Datasets Automatically

Aligning {{glossary.Dataset}}s means placing the datasets at the correct coordinates relative to each other.

Datasets can be aligned automatically, if:

You processed your data in NavVis Cloud and used control points measured with a total station.
You processed your data locally and used control points measured with a total station.
You processed your data locally and used control points that were not measured with a total station.

Important: Locally processed datasets can only be aligned automatically if you used the tools nv_anchor-dataset-alignment and nv_cloudtransform during processing.

If one of these scenarios applies to your data, alignment will happen automatically.

Checking Automatic {{glossary.Dataset}} Alignment

Follow these steps to check automatic dataset alignment if you have processed your data measured control points.

Go to Data Processing >Processed Data.
Select the datasets you want to check.
Click Download.
Select Control point verification log as the output file.
Once the file has downloaded, open it with a notepad application.
Check the third column to find out how successful the alignment was during processing:
- If the difference is <4 cm, your alignment was successful.
- If the difference is 5-6 cm, this may be acceptable for you, depending on your use case.
- If the difference is >6 cm, alignment was not successful.
If the alignment was not as you expected or the differences were greater than 6 cm, process the datasets again and exclude the control points with large differences.

Aligning Datasets Manually

Aligning {{glossary.Dataset}}s is the process of placing the datasets at the correct coordinates relative to each other.

Manual dataset alignment is only necessary for datasets that were captured without control points. Datasets with control points can be aligned automatically. We do not recommend manual dataset alignment for larger projects.

Before You Start

Have the floorplan that you used during your scan to mark the datasets ready.
Check that you have uploaded all the datasets that were scanned and need to be included in the building.
Verify that the dataset is aligned with the building's outline on the street map by clicking on Show Map in Dataset Alignment. If needed, adjust the dataset position manually.

The View Configuration Menu

Top View

Side View (N-S)

The highlighted area is what is visible in the side view. Click and drag the area to move it.

Click and hold down shift to rotate the highlighted area.

Side View (E-W)

The highlighted area is what is visible in the side view. Click and drag the area to move it.

Click and hold down shift to rotate the highlighted area.

Perspective Camera

The perspective camera shows you the angle of the {{glossary.Point cloud}}.

Show Map

If the map does not help you align your datasets, you can choose to hide it.

Selecting a Reference {{glossary.Dataset}}

The reference dataset serves as the starting point for alignment. It is the dataset that you compare the location of other datasets to. Perform the following procedure to select a reference dataset:

Go to {{glossary.Site}} Setup > Dataset Alignment.
Use your floorplan to identify which floor of the building has the most datasets. This should be the floor you start aligning on.
Find a dataset on the floor you have chosen based on the following criteria:
- The dataset overlaps with several other datasets.
- The dataset has features that are visible across multiple stories in the building, such as staircases and elevator shafts.
Once you have identified a suitable dataset, hide all other datasets in the Datasets list by clicking the eye icon next to them, but do not hide the reference dataset.
Set the reference dataset by clicking the star button.
Assign a color to the reference dataset in the color picker.
If it is not already locked, lock the dataset in place by clicking the lock button.

Aligning {{glossary.Dataset}}s on the Same Floor

Note: It is extremely important to save your work. Make sure that you always click Save after any adjustment or alignment.

Select a second dataset on the same floor as the reference dataset. The second dataset must overlap with the reference dataset.

Note: Refer to your floorplan to check which datasets overlapped during scanning. You can then check for matching objects in the overlapping datasets.

Unhide the dataset by clicking the eye icon next to it.
Apply a color to the second dataset that contrasts the color of the reference dataset using the color picker so that you can easily identify the two datasets you are trying to align.
Unlock the second dataset (the dataset you plan to align) by clicking on its lock button.

Note: By default, you should always keep your datasets locked. Only unlock a dataset that you are planning to align.

Click Save.
To align the two datasets, start in the Top View. Place the mouse cursor over the  Top View label and click to bring that view into full screen.

Note: Use the wheel in your mouse to zoom in and out until you can see both datasets on the screen. You can also center directly on a specific dataset by clicking the bullseye button next to the dataset name to bring it into focus.

Drag and rotate the unlocked dataset by using the Transform menu so that the walls of both datasets match. In the images below, you can see what unaligned versus aligned datasets look like from the Top View.
Once the walls match, zoom in and try to match identifying features that overlap between the two datasets such as stairs, windows, doors, elevator shafts, or pillars.
Next, return to the split screen view by clicking on the Top View label to exit the full screen. Then, check the Side Views at the bottom of the screen to see if the floor is also lined up from the sides.
Click one of the Side View labels to enter the full screen and further inspect whether the datasets are roughly aligned. In the images below, you can see what unaligned versus aligned datasets look like from a Side View.
Continue to use the Transform menu to match the datasets in the Top View and two Side Views until the overlapping datasets are roughly aligned in all three views.
Click Save.
Once the datasets are roughly aligned, use the Run automatic alignment function in the sidebar to further align the datasets:
1. Make sure that all datasets except the two you want to align are hidden, that your reference dataset is still locked and the dataset that you want to align is unlocked.
2. Click Run automatic alignment.
3. Click Align horizontally and vertically or Align horizontally only.
4. The system will give you a rough number of the difference between your two aligned datasets. Your automatic alignment was successful if the result is 40 mm or under.
If the walls line up well in the result and the automatic alignment is 40mm or less, click Save and proceed to step 16. If they do not line up or the automatic alignment failed, proceed to step 15.

Note: If you receive an error message, return to each view and adjust the alignment. Then try the automatic alignment again.

If your walls do not line up after automatic alignment, click the Undo button and continue with step 16.
After automatically aligning the datasets either successfully or unsuccessfully, you should continue adjusting the datasets manually as precisely as possible using the Transform menu.
Once the two datasets are aligned to your satisfaction, lock the second dataset by clicking on the lock button.
Click Save.
Change the color of the dataset you just aligned to match the reference dataset using the color picker. This will help you keep track of which datasets you have already aligned during the horizontal alignment and is required for aligning the floors vertically.
Click Save.
Choose another dataset on the same floor that overlaps with the reference dataset and repeat steps 1 to 20.
Once no further datasets overlap with the current reference dataset, choose a new reference dataset on the same floor and repeat the process until all datasets on that floor have been aligned.
Click Publish once you have finished aligning the datasets.

Aligning the {{glossary.Dataset}}s to the Ground Floor

Important: Since you have already aligned your datasets, it is important that you do not adjust this alignment. This step is simply to ensure that your {{glossary.Site Model}} knows where the ground floor is located, so make sure you only move your building as a whole.

Go to Site Setup > Dataset Alignment.
Unhide the aligned datasets that you want to be a part of your {{glossary.Site Model}} by clicking the eye icon next to them.
Use the Side Views at the bottom of your screen to check if the datasets on the ground floor are sitting on the gray line. If it is already aligned, move on to auto-generating your {{glossary.Site Model}}.
If the ground floor datasets are not on the gray line, unlock the whole building by clicking the lock icon next to the building label.
Use the Transformation menu to adjust the alignment of your ground floor so that it is sitting on the gray line.
Use the rotation buttons to rotate the dataset in small increments and/or use the left and right arrows to move the datasets in small increments in the , y, z directions.
Once you have matched your building to the gray line, lock the datasets that you unlocked.
Click Save.
Click Publish once you have finished aligning the datasets.

Aligning the Different Floors of a Building

Choose a dataset on a floor vertically connected to the one you have just aligned (i.e. either above or below). For example, if you aligned the fourth floor, choose a dataset on the third or fifth floor.
Assign the datasets on the new floor the same color using the color picker, so you can clearly identify which floor the datasets belong to.
Unhide all the datasets on both floors by clicking the eye icon next to the floor name.
Unlock only the datasets in the new floor and keep all the datasets in your previously aligned floor locked. For example, if your fourth floor was just aligned, select all datasets on both floors, but only unlock the datasets on the fifth floor.
Start in the Top View. Place the cursor over the  Top View label and click to bring that view into full screen.

Note: Use the wheel in your mouse to zoom in and out until you can see both datasets on the screen. You can also center directly on a specific dataset by clicking the bullseye button next to the dataset name to bring it into focus.

Drag and rotate the unlocked dataset by using the Transform menu so that the walls of both datasets match.
Once the walls match, zoom in and try to find identifying features that overlap between the two datasets, such as stairs, windows, doors, elevator shafts, or pillars. This will help with the Side View alignment.
Click Save.
Go back to the split screen view. Now that you have aligned the floors in the Top View, you must align them in the Side Views so that the floors are properly connected.
To make vertical alignment easier, we recommend using the View Configuration menu to help bring overlapping items into a clearer view.
1. In the View Configuration menu, select either Side View (N -> S) or Side View (E -> W) or both. This will bring a clipping plane into your Top View, which can be adjusted to help you align the floors.
2. Use the rectangular area that appears in the Top View panel to adjust the content visible in the Side View below, depending on which side you have selected. Do this by moving your mouse over the rectangle and scrolling out with your mouse to adjust the size. Drag and drop the rectangle to where you would like it to be placed.

Note: You can rotate the view area by moving the rectangle on the top left panel while holding the Shift key.

Once you have brought the object you want to align into focus, go into one of the Side Views to get a better picture of the floors by clicking on the Side View label.
Using the Transform menu, move the unlocked dataset to connect it to the locked datasets.
Click Save.
Double check in the other Side View that the floors are properly aligned. Then go back to the main split screen overview.

Note: To ensure that each story of the building is correctly vertically aligned, try to make sure there is a small gap between the floor of one dataset and the ceiling of the dataset below it. If this is not possible, it is better to cut off the ceiling instead of the floor.

Once you have aligned the datasets, lock the datasets you moved.
Click Save.
Once you have finished aligning one floor to another, use the Top View again to check for alignment errors on the floor that you just added.
Start in the middle and check that overlapping features between floors like staircases or windows match.
Move to the building outline and check the ends, wings and outer edges. Outlines between floors should match each other. If that is not possible, try to distribute the error instead of concentrating it in one part of the building.
Click Save.
If you find areas that are not aligned, choose a new reference dataset and repeat the necessary steps from Aligning datasets on the same floor.
Check the overlapping datasets on that floor for alignment. Use the Transform menu to make any adjustments.
Once you are satisfied with the alignment of these two floors, repeat the entire process for aligning floors to one another with the next set of datasets. For example, if you aligned the fourth and fifth floors, now align the sixth floor.
Click Publish once you have finished aligning the datasets.

Placing Aligned {{glossary.Dataset}}s on the Map

If you did not already place your site on the world map through the use of site coordinates, the final step in dataset alignment is to make sure your aligned building is in the correct location.

Select all datasets.
Go into the Top view.
If the Top view does not show a map of the site, select Show map in the View Configuration menu.
To help you find the right location for your datasets, you can show the base point of the site by clicking the eye icon next to Site Base Point.
Unlock all datasets by clicking the lock icon next to Site Base Point.
A bounding box will appear around your datasets.
Use the bounding box to drag the datasets to the correct place. If necessary, use the rotation arrow to move the datasets into the correct angle.
Click Save.
Click Publish once you have finished aligning the datasets.

FAQ

What does aligning datasets mean?

Aligning datasets means placing the datasets at the correct coordinates relative to each other.

Can datasets be aligned automatically?

Yes, datasets can be aligned automatically if specific conditions are met, such as using control points measured with a total station.

What tools are required for local dataset alignment?

For locally processed datasets, the tools 'nv_anchor-dataset-alignment' and 'nv_cloudtransform' must be used during processing for automatic alignment.

How can I check the success of automatic dataset alignment?

You can check the success of automatic dataset alignment by downloading the Control point verification log and reviewing the third column for alignment differences.

Is manual dataset alignment recommended for larger projects?

No, manual dataset alignment is not recommended for larger projects; it is only necessary for datasets captured without control points.

Generating Maps