FaceDetect Infirence

I don’t really understand how to interpreter/postproses output of the facenet (detectNet)

The shape is 46x26x4 4 is (xc, yc, w, h) but what is 46 and 26? one correspond to class and another is corresponding value of box or what? ¯_(ツ)_/¯

This looks like TAO related, we are moving this post to the TAO forum to get better help.

See DetectNet_v2 — TAO Toolkit 3.22.05 documentation
DetectNet_v2 generates 2 tensors, cov and bbox. The image is divided into 16x16 grid cells.

So 736x416 ==> 46x26

There is no update from you for a period, assuming this is not an issue anymore.
Hence we are closing this topic. If need further support, please open a new one.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.