Forum

Author Topic: Confidence math  (Read 2176 times)

glennk

• Newbie
• Posts: 15
Confidence math
« on: April 05, 2022, 06:28:27 PM »
Hello,

I can't find any documentation on how Metashape's confidence is computed.
It is a value in [0, 255] in the GUI, in Python the range is [0, 33].

This thread does not help much since I looking for an idea of the math behind confidence.

Can someone explain me a little more in depth what really is confidence ?

I am using this value for research purposes, that's why I would like to know more about it.

Best regards,

Glenn

Alexey Pasumansky

• Agisoft Technical Support
• Hero Member
• Posts: 14965
Re: Confidence math
« Reply #1 on: April 06, 2022, 12:15:52 PM »
Hello Glenn,

Do you mean dense cloud points confidence or mesh vertices confidence? Or both?
Best regards,
Alexey Pasumansky,
Agisoft LLC

glennk

• Newbie
• Posts: 15
Re: Confidence math
« Reply #2 on: April 07, 2022, 03:55:22 PM »
Hello Alexey,

Thank you for the quick reply

I do not know the difference between dense cloud points confidence or mesh vertices confidence, so I guess I mean both.

Best regards,

Glenn

Alexey Pasumansky

• Agisoft Technical Support
• Hero Member
• Posts: 14965
Re: Confidence math
« Reply #3 on: April 07, 2022, 08:14:17 PM »
Hello Glenn,

For the dense cloud points the confidence value means the number of the depth maps involved to the point generation process. The value is integer.

The confidence value for the mesh vertices carries the statistical information about the averaged number of depth maps used for the model reconstruction in the neighborhood of the given vertex. The value is float.
Best regards,
Alexey Pasumansky,
Agisoft LLC

c-r-o-n-o-s

• Jr. Member
• Posts: 91
Re: Confidence math
« Reply #4 on: April 07, 2022, 09:44:39 PM »
Hello Glenn,

For the dense cloud points the confidence value means the number of the depth maps involved to the point generation process. The value is integer.

The confidence value for the mesh vertices carries the statistical information about the averaged number of depth maps used for the model reconstruction in the neighborhood of the given vertex. The value is float.

Thank you!

glennk

• Newbie
• Posts: 15
Re: Confidence math
« Reply #5 on: April 08, 2022, 10:54:30 AM »
Hello Alexey,

This is very clear and helpful, thank you.

Best regards,

Glenn

gautier.laurent

• Newbie
• Posts: 2
Re: Confidence math
« Reply #6 on: December 13, 2022, 05:37:46 PM »
Agreed.
I have been looking for this information for quite a while and I'm needing it to clarify results for research.
So many thanks for asking the question here and for the clear answer.

I'd suggest to clearly integrate this information in the manual,
especially considering how simple the answer is in the end.

I was suspecting the math would be complicated and that was why it is not explained.