Interpreting Depth Confidence

Software
1

ZED SDK Version 4.1.3 [Latest]4

Hey guys, I have a question on interpreting Depth Confidence values. In the docs:
“It gives every pixel (X, Y) in the image a value in the range [1,100], pixels having a value close to 100 are not to be trusted.”

It says that the range is [1,100] inclusive, but when I retrieve the confidence measure, The numpy array contains values below 1. I am using NEURAL_PLUS. Is this intended behaviour, and does it change the above interpretation of depth confidence?

2

Hi @mukundbala,

This is a typo, and the confidence range is indeed [0, 100].

Will update the docs with this, it does not change the behavior or interpretation of the confidence map.

1 Like
3

I see, we’ve been clipping values to [1,100]. I doubt this will affect anything, but good to know! Thank you for your hard work :slight_smile: