131–139 of 139 posts
http://i.imgur.com/Lq648My.png Well, there goes my last ounce of self esteem. If anyone needs me I'll be sitting in the corner with one of those criminal hacker ski masks while I work on open source stuff.
Feel better about yourself: http://imgur.com/uJVnQ9S
On a more serious note, training set probably didn't include data from Asia.
Edit: Just scrolled down and saw your question (and the answer you received). :C
Earlier quoted context omitted.
As an example (both me, and I'm in my early 30s): 24, "Hmm...": http://imgur.com/V1AMDfd 27, "Stunning": http://imgur.com/bmmJ2dg I'm actually a year younger in the second picture than the first. Maybe it's the glasses?
Any chance of seeing a photo of you in guymode that it recognises as being in guymode? In its defence, you do look very different in those photos. I wouldn't recognise you as the same person. That said, I'm surprised that it ranked you so poorly in the first photo.
27/Hot: http://imgur.com/mHpvd5j 36/Nice: http://imgur.com/a2IKkRD 30/Ok: http://imgur.com/Vicvbq3
First is a year and half old, the other two are within the last couple months.
Earlier quoted context omitted.
Of course I tried a bunch of pictures of myself, just as everybody at HN is doing, ha ha! I didn't get above the nice level. First picture of my wife she immediately got the godlike level. Unfair advantage!! "Before performing any experiments we removed underage people, anyone over 37 and bi- and homosexual users as these comprise only a small minority of the dataset." That's already biasing it of course! "Interestin…
Yes, this was the data we worked with for attractiveness modeling. The algorithm learns from millions of ratings from males to females and from females to males, all heterosexuals. No underage or above 37 years old, and mostly people from Zurich area, Switzerland. For age and gender prediction we use more and diverse data. I am Radu Timofte, one of the authors.
Is it possible that quite a few men rate every women high irrespective of looks?
Or are women just more picky? Why is it called biased?
When I put a picture of myself that had the contrast turned up my picture was rated "godlike", when I put in the original picture it was rated "hot", and when I turned down the contrast some it was rated "ok." I'm an unhealthy, almost pasty shade of white with a slightly bulbous nose and a classic fivehead. Now, this a problem for bigger reasons: An older picture of Denzel Washington gets an ok: http://imgur.com/Li0g…
If people are not mature enough to understand that a machine learning algorithm cannot be racist, shouldn't you go educate them instead of telling people not to offend these idiots? Secondly what makes you think that every race is equally attractive universally? Studies have people find people of their own race more attractive. If whites are more attractive (either by their training set or user base ratings) doesn't…
Earlier quoted context omitted.
Any chance of seeing a photo of you in guymode that it recognises as being in guymode? In its defence, you do look very different in those photos. I wouldn't recognise you as the same person. That said, I'm surprised that it ranked you so poorly in the first photo.
Sure. 27/Hot: http://imgur.com/mHpvd5j 36/Nice: http://imgur.com/a2IKkRD 30/Ok: http://imgur.com/Vicvbq3 First is a year and half old, the other two are within the last couple months.
That said, I think I'd agree with its ordering of those photos. shrug
Earlier quoted context omitted.
Yes, this was the data we worked with for attractiveness modeling. The algorithm learns from millions of ratings from males to females and from females to males, all heterosexuals. No underage or above 37 years old, and mostly people from Zurich area, Switzerland. For age and gender prediction we use more and diverse data. I am Radu Timofte, one of the authors.
What did you mean about the bias of the rating by women? Is it possible that quite a few men rate every women high irrespective of looks? Or are women just more picky? Why is it called biased?
It's not a lot more precise than randomly guessing. Check this album out: http://imgur.com/a/1a1tn Seems racist and sexist towards men.
Our data consists only from normal (or natural looking) face images in the wild (from IMDB, Wiki, and/or BLINQ user profiles). On such data we get very good apparent age prediction (better than the human reference) and also very good gender and attractiveness prediction.
The attractiveness is highly subjective and its perception varies from one culture/region to another. We used data from Switzerland.
Our solutions are far from being perfect and the guessed results should not be taken too seriously.
We consider to update our models to explicitly deal with distortions and non-human face contents.