This forum is disabled, please visit https://forum.opencv.org
 | 1 | initial version |
you can set the threshold value from the create function, e.g.here
to find a good number there, you have to experiment, collect the confidence values from predict() for a few "same" and "not-same" person values, and pick a value between the worst "same" and the best "not-same" confidence (which is actually a distance)
| | 2 | No.2 Revision |
you can set the threshold value from the create function, e.g.here
to find a good number there, you have to experiment, collect the confidence values from predict() for a few "same" and "not-same" person values, and pick a value between the worst "same" and the best "not-same" confidence (which is actually a distance)distance, so smaller==better)
