Slashdot Mirror
Women's Attractiveness Judged by Software
Roland Piquepaille writes "According to Haaretz, an Israeli team of computer scientists has developed software that ranks facial attractiveness of women. Instead of identifying basic facial characteristics, this software has been designed to make aesthetic judgments — after training. The lead researcher said this program 'constitutes a substantial advance in the development of artificial intelligence.' It is interesting to note that the researchers focused on women only. Apparently, men' faces are more difficult to grade."
There are some obvious criticisms: In the first stage, 30 human participants were asked to rate from 1-7 the beauty of several dozen pictures. For a masters project (which this was), that's a decent sample size. For research and practice, I do not think that will suffice.
Second, this was done using eigenalysis and principle component analysis. While that's interesting, I have not always found that to be a great approach. Five or six years ago, they were all the rage although I cannot really find anything fruitful that has come from applying this to human faces. This also means that they cannot generate the 'most beautiful' face but if they did, it would simply be the composition of all their eigenvectors (in this case, ghostly looking images of faces) into one representing the highest scoring beauty. The lead researcher said this program 'constitutes a substantial advance in the development of artificial intelligence.' Having taken several AI, computer vision & machine learning courses, I don't find this to be at all substantial. An interesting masters project for sure, but several years ago I saw people doing the same things at local universities with the same results.
Why don't they tell us how this scored some celebrities from around the world like say Iman Abdulmajid, Zsa Zsa Gabor & Angelina Jolie? I have a feeling that their system is over-trained and would perform poorly in real life. Facial beauty requires imagination and this system was hand trained on a hundred points. I don't think that's enough but I wish they would have published more results to either prove or disprove my criticisms.
My work here is dung.
"Apparently, men' faces are more difficult to grade."
Or perhaps their bank accounts are easier to derive a "value" from!
I kid, I kid. I think.
Men will nail anything and the women really control sexual interactions. The cost of mating is far lower for men than for women therefore women are far more choosy.
Engineering is the art of compromise.
The idea that there are no set standards for beauty is wishful thinking. Every guy wants to find a girl who is beautiful and for some reason nobody else has noticed. In reality this never happens. The next time you see a pretty woman in the airport, don't look at her, look at all the guys as she walks by, it's quite noticeable. Attractive people are treated better from a young age and, knowingly or unknowingly, they leverage this asset to get what they want. This is not some quirk in the study of psychology, it's the driving force behind the behaviors that shape evolution. It's a cruel trick of nature that we are not all created equal, and I'm glad we're civilized enough to moderate some of the resulting inequality.
I fell in love, unexpectedly, with a woman who was not a classic beauty.
Within a handful of months, I noticed I was finding women with facial and body characteristics similar to hers more attractive than the magazine beauties I normally ogled. Indeed, the model types started looking odd to me.
Now add in cultural and racial preferences and this "breakthrough" starts sounding like "bullshit".