what you find below is a summary of 40 pt. computer program for a beautiful face. the program presents diverse fixed facial points "ratios." the idea being that the program expresses real beauty.
by "real" the programmers mean these ratios yield an objective ideal (obviously, the larger the set the better the approximation).
This work presents a novel study of the notion of facial attractiveness in a machine learning context. To this end, we collected human beauty ratings for data sets of facial images and used various techniques for learning the attractiveness of a face. The trained predictor achieves a significant cor- relation of 0.65 with the average human ratings. The results clearly show that facial beauty is a universal concept that a machine can learn. Analysis of the accuracy of the beauty prediction machine as a function of the size of the training data indicates that a machine producing human-like attractiveness rating could be obtained given a moderately larger data set.the programmers make a point about "real" beauty, as a concept, i.e., quantifiable. given a discreet number of variations, the program yields a "standard" of shared attributes. each sample (face) of the set is already picked as representative of its kind.
here is the 40 item-point list (we abbreviated for the sake of space):
1. Face length
2. Face width—at eye level
3. Face width—at mouth level
4. Distance between pupils
20. Nose width at nostrils
21. Nose length
22. Nose size = width * length
25. Thickness of middle of top lip
26. Thickness of right side of top lip
27. Thickness of left side of top lip
38. Ratio of (distance from nostrils to eyebrow top) to (distance from face bottom to nostrils)
39. Skin smoothness indicator (description follows)
40. Hair color indicator (description follows)
do you agree? if not, why not?