Using Draft to establish Alignment settings

Development Asset Creation
,
1

When I use “Draft” settings to quickly find out better Alignment settings, I can only use 3 factors, right?
It’s only “Images overlap”, “Image downscale factor” and “Final model optimization”?
The other settings don’t contribute to it, right?

When I need more improvements (for example when I have many dis-connected components), I have to use control points, right?

Thank you.

Edit:

The video preview can perfectly fit the many groups together, I don’t see any lap.
Is there a simple way to combine these groups?

2

Hello tw2016.

You can read RealityCapture manual “Alignment Settings”.

Max features per mpx

  • Max features per image
  • Images overlap defines
    Image downscale factor
  • Max feature reprojection error
    Advanced:
    Force component rematch
    Background feature detection
    Background thread priorit
  • Preselector features
  • Detector sensitivity low / medium / high / ultra
    Merge components only
    Merge georeferenced components
  • Distortion model

(+) is most important for alignments.

3

Thank you. You don’t use the Draft mode to find out the best alignment settings, but you play with the settings that you have mentioned, and you run it in “Normal” (Not Draft) mode, right?

4

tw2016 wrote:

Thank you. You don’t use the Draft mode to find out the best alignment settings, but you play with the settings that you have mentioned, and you run it in “Normal” (Not Draft) mode, right?

Hello tw2016,
the draft mode is meant as a quick preview of the result of your input data - most useful in the field - so that you alternatively add more of them or do some adjustements in your capturing methods.