Using GCPs in XYZ format


I’m new to Web ODM and I work in car accident reconstruction. I have created orthos and point clouds from two of my cases so far using WebODM just to teach myself how to use it but I’m having trouble using the GCP interface.

I use a robotic total station so when I map the GCPs, they are not in Lat/Long, they are in XYZ format from an arbitrary point (the location of the total station).

Is there a way to use XYZ formatted GCPs in Web ODM? Below is the text file from the GCPs that I exported from my CAD program (Trimble Reveal):

6, -119.068, -11.034, -0.130, 0.098, 0.098
7, -34.280, -21.433, 1.222, 0.098, 0.098
8, 22.077, 6.437, -0.739, 0.098, 0.098
9, 4.768, 44.702, -0.214, 0.098, 0.098

Car accident scenes are relatively small and I have (in the past) just used 5 Control Points…I was also using GNSS mapping equipment and Pix4D at the time. I am now in private practice and I only have the total station

Appreciate any help,

1 Like

As far as I understand, they’d need to be converted from offsets to absolute position in some CRS.

If you have the exact position of the TotalStation, this is doable, though likely not super fun.

1 Like

Ideally yes… but if they really don’t care about CRS location, then using those relative coordinates and calling it some arbitrary planar CRS (preferably a UTM) and using GCPs would be just fine. Just know it’s going to show up on the map in some random location.


Can you elaborate on how I could do this? I’m trying to figure out if webodm can use an arbitrary coordinate system.

Can you suggest some sort of workflow I could use to convert the GCP format above to a UTM (even arbitrary) format that webodm could recognize?

Hi Smokey,

I think a Total Station is more than enough for what you are trying to do. And 5 control points is OK.

As for the coordinate system you can use UTM instead of Geographic Coordinates and get the stating point for your total station using a handheld GPS or a Smartphone. Then orienting the total station using the tubular compass and adjusting the magnetic declination acording to your location.

This method will give you high relative accuracy and low absolute accuracy. In other words, the model will be accurate but misplaced in the real world because of the accuracy of the Handheld GPS which is tipically 3 - 5 meters.

I hope this information can help you.


Excellent as always, Isreal.

If you have Android and a bit of patience, GPS Point/Position Averaging (long-baseline sampling) goes a LONG way towards improving over standard single-sample accuracy.

I like this app:
GPS Point - Apps on Google Play

Sitting on my desk indoors in my apartment, no clear view of the sky from an entry/mid-tier MediaTek SOC Phone:

That is to say, you can do much better with applying a few simple GPS collection methodology things you’d normally do.


Ahh, that might be the solution there.

It is relative accuracy that I need, thanks very much.


Let us know how you get on!

1 Like

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.