Yes, I found tagging a person less than about ten times tends to mess up the recognition. The sweet spot is about 20-50, then as described above as soon as you have a couple hundred for one person it seems to overfit. I could identify half the detected faces pretty easily by re-running detection two or three times but eventually I ended up with a remaining set of faces that always got detected as the wrong person, if at all, and it's not that they are all blurry or otherwise low quality.
Also I wonder how bad it is to accidentally assign the wrong tag to a person. I noticed that it happened three or four times, so I ended up having maybe 400 proper assignments and then one false tag. Could they even add a function to detect this?
The next thing for me would be to research which management tools support face recognition and what they use to do it. Anything using what digikam uses would not be worth trying to me.
Also I wonder how bad it is to accidentally assign the wrong tag to a person. I noticed that it happened three or four times, so I ended up having maybe 400 proper assignments and then one false tag. Could they even add a function to detect this?
The next thing for me would be to research which management tools support face recognition and what they use to do it. Anything using what digikam uses would not be worth trying to me.