Sorry for opening a new topic, but I have another issue. When I align the laser scans and photos, I set the laser scanner scans to a locked position and use a local reference system. However, when I align them with the photos, everything gets slightly tilted, causing a significant error. Could I get some explanations on how to lock the position of the scans without it being changed During the alignment phase?
Hi, during the import of the laser scans, use the following settings:
To preserve the imported coordinate systems of the laser scans during export from RC, make sure the Project and Output coordinate systems in the Application settings are set to local:1 - Euclidian as well.
Finally, when you export the result from RC, make sure the Coordinate system is set to Project output. This will preserve the laser scans coordinate system throughout the whole RC pipeline.
Are the photos you are trying to align to the laser scans georeferenced? If so, this might be causing issues and big errors. If yes, disable the georeferencing in the Prior pose. Simply select all photos and set the absolute pose to Unknown. Only the laser scans will have a locked prior pose.
I recommend you delete all components and redo the alignment from scratch. RC tries to optimize existing components, and if they are already bad, it’s better to start from scratch.
Let me know if this advice helps.
I think the problem was either with or without the georeferencing. Now i choose yes and the model are correct. thanks
And what do these parameters refer to
The problem persists even if I insert the scans as georeferenced in the exact position and lock them. You can see the parameters below. The problem persists when I use the photos… why
Hi @geoanture86
Lock for continue option locks only the relative poses of the cameras, not the absolute ones.
Locked pose group - there you can set the camera groups which relative poses won’t change after alignment.
Don’t lock laser scans. It is not necessary, as you are using the exact option during import.
Also, keep Yes for Use camera priors for georeferencing and set Unknow just for images in 1Ds view for Prior pose.
Thank you, but I didn’t understand why the first time, even though I set the exact position and locked the laser scans, after the alignment, the whole component was tilted. How can you explain this to me
Hi, following your settings it is because you have set No under Use camera priors for georeferencing. To keep the laser’s position and orientation you need to set it to Yes.
For other images you can set the Prior pose to Unknown.
Now understand…
i have another question in the menù con.
can i write a code? if yes, which is the language?
I noticed that when the alignment finishes, it performs a color correction. Why is that?
I have three sets: the first set is a nadir drone mission georeferenced using GCPs, a set of photos taken with a drone and therefore with GPS position, and a laser scanner set georeferenced in another program using the same GCP coordinates. Now, I want to merge them into one set. What do you recommend I do?
can i write a code? if yes, which is the language?
You can use batch scripting or sequencer: RealityCapture Help, https://www.youtube.com/watch?v=cZv_DKQMSVg
I noticed that when the alignment finishes, it performs a color correction. Why is that?
You need to have set that in the Color and Texture settings:
I have three sets: the first set is a nadir drone mission georeferenced using GCPs, a set of photos taken with a drone and therefore with GPS position, and a laser scanner set georeferenced in another program using the same GCP coordinates. Now, I want to merge them into one set. What do you recommend I do?
I suppose you asked this more times before. Jakub described the process above, the process was also described here: georeferencing and registration a component - #6 by OndrejTrhan.
If all your datasets are georeferenced, you can import the components into new project and then use the option Merge georeferenced components in the Alignment settings.
Just drone images could have not so precise GPS, so I would use the GCPs also there.
Thanks
Hi, I aligned my first set of photos with the GCPs in one project.
I aligned the second set of photos in another project using some of the same GCPs as before.
I exported the components and reimported them into a new project, I did a merge, but it didn’t give me any result
Hi, just to be sure, do you still describing the same project?
As you wrote that you used the GCPs to georeference your components, why they are not in the project? After exporting the component and then importing it into a new project, the control points are imported, too.
According to your image it looks just like imported components into a new project.
By merge, do you mean a tool Merge Components?
Have you tried a new alignment in this project?
Also, as not of your components are georeferenced and if the alignment wont’ work, you can set Merge georeferenced component to Yes in the Alignment settings and then align again.
"I georeferenced the two components in two different projects. When I imported them into the new project, I deleted the GCPs and redid the alignment, also using the merge component function in the alignment section.
Why you didn’t keep those GCPs? It will help you to merge the datasets and also keep scale and position.
It is strange that for such overlap between components you are getting two of them using the alignment…
Can you try also that Merge georeferenced components setting?
Which GCPs should I use, from the first project, the second project, or both?
I tried with the ‘Merge georeferenced component’ option, but nothing worked.
Now, for the two components, I used some GCPs and tried to align them, but there’s a significant error in altitude.
Can you explain the best way to align these components?
Consider that the second mission has detailed images of the object, so it didn’t capture all the GCPs present in the nadir mission.
This is what I’m doing now: using the first georeferenced nadir project, I import all the detailed photos from the second mission and start the alignment. If it doesn’t work, I’ll also add the GCPs I can find in the photos from the second mission
You wrote that the GCPs were the same for both project, so it doesn’t matter…
As the components are already georeferenced, that option should work.
As you used the GCPs again, have you turned off the prior pose for the images?
You don’t need to capture all of the GCPs, for a basic alignment you need to have at least three. How many GCPs do you have in your project? How are they placed?
If GCPs doesn’t work, I would use also control point to merge the components.
I used 14 GCPs for the first project and 5 GCPs for the second one. I didn’t deactivate the prior pose.
Can you explain what the prior pose is used for and when to activate it? I still don’t fully understand.
