At what step are "ground control points" or markers taken into account?

[nanders]

Hi all,

short question you may know immediately:
In what step in the workflow are the markers (ground control points) taken into account?

For example: I have finished an entire workflow, and afterwards I added reference points. Do I need to recalculate the depth maps or can I start generating the mesh from existing depth maps? (much quicker)

thanks,
Niels

[jedfrechette]

I've been trying to sort this out too. Assuming I have independently measured ground control points and I want to use guided marker placement is the correct workflow:

• Align photos
• Build geometry (low-res for guided placement)
• Place markers
• Optimize camera parameters & point coordinates
• Build geometry (including depth maps for final model)

Although only the intrinsic camera parameters are listed I'm assuming the optimize step also optimizes the extrinsic parameters. Before and after inspection of the camera transform matrixes suggests this is the case and it certainly seems like the right thing to do. I'm also not clear on how the "Update" function on the Ground Control panel integrates into this?

Step 5 is necessary because all of the camera parameters have changed since the geometry was originally built in step 2.