16
Camera Calibration / Re: Adaptive camera model or calibration groups
« on: October 20, 2021, 10:07:59 AM »
...split in two. That is for sure.
Show Posts
This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.
17
General / Re: Added Erosion radius option to Classify Ground Points dialog
« on: October 13, 2021, 08:55:10 PM »
That is of course a very good approach!
For the classification, I still prefer the approach where the terrain is evaluated "from below".
This way there are no "stumps".
I often classify terrain on slopes, where Metashape has problems, but I'm going to try out the new version.
For the classification, I still prefer the approach where the terrain is evaluated "from below".
This way there are no "stumps".
I often classify terrain on slopes, where Metashape has problems, but I'm going to try out the new version.
18
General / Re: Added Erosion radius option to Classify Ground Points dialog
« on: October 13, 2021, 11:45:48 AM »
*irony on*
Have a look at the manual.
*irony off*
(Such software with so many complex settings really deserves better documentation.)
Have a look at the manual.
*irony off*
(Such software with so many complex settings really deserves better documentation.)
19
Python and Java API / Re: Scipt to turn on/off GPU
« on: October 06, 2021, 01:10:24 PM »
So, here is a result from my "quick and dirty" test.
It was not the optimal PC and project for this, but without GPU the point cloud is calculated about 2% faster.
( Average of 4 calculations with and without GPU. )
So really a small "overhead" but certainly negligible.
For very large projects, however, this can certainly add up to a few hours.
It was not the optimal PC and project for this, but without GPU the point cloud is calculated about 2% faster.
( Average of 4 calculations with and without GPU. )
So really a small "overhead" but certainly negligible.
For very large projects, however, this can certainly add up to a few hours.
20
Python and Java API / Re: Scipt to turn on/off GPU
« on: October 06, 2021, 10:23:04 AM »
Hello.
In the table from the link you can read that the generation of the dense point cloud and the creation of an orthomosaic with 2 graphics cards takes a little longer.
(However, this refers to version 1.6.5. I will do my own tests with the latest version in the next few days. )
If this is the case, you could first create the DEM with depth images and 2 graphics cards in a stack.
Then comes a script in the batch that turns off one graphics card.
Then the point cloud and the orthomosaic can be calculated with one graphics card.
In the table from the link you can read that the generation of the dense point cloud and the creation of an orthomosaic with 2 graphics cards takes a little longer.
(However, this refers to version 1.6.5. I will do my own tests with the latest version in the next few days. )
If this is the case, you could first create the DEM with depth images and 2 graphics cards in a stack.
Then comes a script in the batch that turns off one graphics card.
Then the point cloud and the orthomosaic can be calculated with one graphics card.
21
Python and Java API / Scipt to turn on/off GPU
« on: October 04, 2021, 07:37:32 PM »
Hello.
I don't know if this overview is still up to date:
https://www.pugetsystems.com/labs/articles/Agisoft-Metashape-1-6-5---NVIDIA-GeForce-RTX-30-Series-Multi-GPU-Performance-2009/
At least in the 1.6.5 version it seems that in some areas the performance is better with just one GPU.
Could a graphics card be switched on/off via batch script for the specific task?
THX
I don't know if this overview is still up to date:
https://www.pugetsystems.com/labs/articles/Agisoft-Metashape-1-6-5---NVIDIA-GeForce-RTX-30-Series-Multi-GPU-Performance-2009/
At least in the 1.6.5 version it seems that in some areas the performance is better with just one GPU.
Could a graphics card be switched on/off via batch script for the specific task?
THX
22
General / Re: Import German Geoid Model into Metashape
« on: September 16, 2021, 02:09:13 PM »Quote
It would be a lot easier if I could define a separate geoid for each survey project with a separate undulation per project.
Of course it is possible...
(It says so on page 5 of my instructions).
Instead of "Add", enter "User-defined".
A fixed value can be entered there.
As already written, the lengths should not be too long.
(In NRW up to 4cm undulation change per 1km).
But then you can also stay in eleptoid heights and change the coordinates with a "Z-OFFSET" script.
(These scripts can be found at Agisoft).
https://github.com/agisoft-llc/metashape-scripts/blob/master/src/add_altitude_to_reference.py
23
General / Re: Import German Geoid Model into Metashape
« on: September 15, 2021, 01:18:31 PM »
The data from this page definitely works.
(Germany/NRW free registration)
https://www.bezreg-koeln.nrw.de/brk_internet/geobasis/geodaetische_fachprogramme/transformation/echtzeit/index.html
Instructions on how to set it up should be included, or I can provide them.
You have to be careful with the non-TIF formats.
There was a bug before V 1.7 that the files were interpreted as mirror images.
Offsetting geoid heights does not work over a "long distance".
Depending on the length of the area, the deviation then goes into the cm range.
(I have an example where 4cm of undulations can be seen over a 1km length.)
(Germany/NRW free registration)
https://www.bezreg-koeln.nrw.de/brk_internet/geobasis/geodaetische_fachprogramme/transformation/echtzeit/index.html
Instructions on how to set it up should be included, or I can provide them.
You have to be careful with the non-TIF formats.
There was a bug before V 1.7 that the files were interpreted as mirror images.
Offsetting geoid heights does not work over a "long distance".
Depending on the length of the area, the deviation then goes into the cm range.
(I have an example where 4cm of undulations can be seen over a 1km length.)
24
General / Re: Import German Geoid Model into Metashape
« on: September 14, 2021, 10:47:09 AM »
Small side note...
Do not use the GeobasisV01.tif file. (from NRW)
According to my verification, this file was faulty. There is now a new and official one on the "Geobasis NRW" homepage.
@Dieter
Melde Dich per PM bei mir wenn es weiter nicht funktioniert.
Do not use the GeobasisV01.tif file. (from NRW)
According to my verification, this file was faulty. There is now a new and official one on the "Geobasis NRW" homepage.
@Dieter
Melde Dich per PM bei mir wenn es weiter nicht funktioniert.
25
Bug Reports / Re: 1.7.4 Point Cloud Variance
« on: September 03, 2021, 02:45:59 PM »
I have now opened the project from the previous version 1.7.0.
No variance is displayed here either.
No variance is displayed here either.
26
Bug Reports / Re: 1.7.4 Point Cloud Variance
« on: September 02, 2021, 10:55:16 PM »
That's how it is, that's how it happens.
But I don't think it is with every camera.
But I don't think it is with every camera.
27
Bug Reports / Re: 1.7.4 Point Cloud Variance
« on: September 02, 2021, 08:12:46 PM »
...of course. 
That was not the problem.
I think it's because of that:
I did the camera optimisation very often and always deleted more cameras.
At a certain point, however, the display simply stopped coming up.
Only when the camera alignment was completely reset did it work again.
Definitely a "small" bug.
If I can repeat it more often, I could make the project available...
That was not the problem.
I think it's because of that:
I did the camera optimisation very often and always deleted more cameras.
At a certain point, however, the display simply stopped coming up.
Only when the camera alignment was completely reset did it work again.
Definitely a "small" bug.
If I can repeat it more often, I could make the project available...
28
Bug Reports / 1.7.4 Point Cloud Variance
« on: September 02, 2021, 04:52:34 PM »
Hello.
Since the new version, the display of the variance does not work for me very often, but not always.
The selection works via "Gradual Selection", but the display (red, yellow, blue) is not shown.
The icon for displaying the point cloud variance is also greyed out.
Can someone test this?
(As I said, it is not always...)
Since the new version, the display of the variance does not work for me very often, but not always.
The selection works via "Gradual Selection", but the display (red, yellow, blue) is not shown.
The icon for displaying the point cloud variance is also greyed out.
Can someone test this?
(As I said, it is not always...)
29
General / Re: Ground Classification Algorithm
« on: September 01, 2021, 10:53:00 PM »
You cannot compare them.
CloudCompare can use the other approach, that the terrain is seen from "below".
The results are also only good if the parameters are correct.
CloudCompare can use the other approach, that the terrain is seen from "below".
The results are also only good if the parameters are correct.
30
General / Re: Image alignment processing times 1.7.3 vs 1.7.4
« on: August 18, 2021, 10:59:04 AM »
That's very interesting.
Did you run the tests twice?
Laptops are already quite sensitive to heat and regulate down faster.
Of course, that's probably not the reason...
Did you run the tests twice?
Laptops are already quite sensitive to heat and regulate down faster.
Of course, that's probably not the reason...
