Live data from Hacker News

Let artificial intelligence guess your attractiveness and age

faces.ethz.ch

91–100 of 139 posts

Re: Let artificial intelligence guess your attractiveness and age

#91
post #81

Earlier quoted context omitted.

31-year-old trans MTF here... uploaded my most recent picture, and it said I was 20, female, and ice cold "Hmm...". Uploaded a picture from about a month ago... same, except it said I was 19. Well, I'm flattered it thinks I can't drink, and I'm glad I pass. Too bad it doesn't think I'm hot, but I've always preferred to go for cute over hot anyway. I'm gonna dig through my photos and see how consistent this is... (edi…

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?

I think the angle might also help. Your jawline is much less prominent in the second one, so that might be why it's "Stunning" rather than "Hmm...".

Here are the ones I used: http://imgur.com/a/yFwN4 -- the first one was taken today, the rest are all from the last couple of months.

Re: Let artificial intelligence guess your attractiveness and age

#92

Earlier quoted context omitted.

Get rid of the headphones and try again

At the time that picture was taken, I didn't own speakers. Now, I look even worse and don't own a working webcam.

How can you look even worse? That picture is terrible, a 1 or a 2 out of 10. Anyway here's a plan for you:

Plan A) if you have any amount of disposible income (more than $0):

- spend 15 minutes reading about beginning to lift weights if you've never lifted a weight in your life

- go to the gym once (ever) i.e. on a day pass.

This plan I've listed takes 90 minutes if you start right now - as in 90 minutes from now you can be back home - and is 97% of what it takes to spend 20 minutes in a gym once or twice per week. if you've found a gym in your area and have tried it out, you're 97% of the way there. For a man, it takes basically no time whatsoever to become fit, it's literally a matter of reading about it and going there, figuring out that gyms exist. In terms of time commitment it is basically zero - in fact, because you have more energy and get things done when you're back, it's negative time commitment. You have more time in your life if you go to the gym than if you don't. Try it out and then sign up if it works. you can be back home within two hours.

Plan B) since you mention not owning speakers and not owning a working webcam, perhaps you have $0 disposible income.

- In this case begin each day with a gradually increasing number of pushups, situps, and you can do these kind of dips - https://www.google.com/search?q=chair+dip (I've listed three exercises.)

Just make it part of your day. You won't become very buff, because all of the mentioned exercises have high mechanical advantage: the reason plan A mentions the gym is because you won't find heavy weights at home and shouldn't bother trying. Home exercises aren't great. they'll get you into basic form though.

Like plan A, plan B also takes negative time (you have more time in the day - because you'll be pumped and ready to get things done - than if you didn't do it), and in terms of wall time takes maybe 90 seconds. Do it right when you wake up.

Finally, if you ever have 4 weeks where you can look scruffy and homeless, as a project you can grow a beard (simply stop shaving, and see what happens after 4 weeks), and then go to a proper barber shop to have them trim and style it for you, tell them you're good with any style and have grown a beard for the first time ever, that they should pick one for you. This will add +3 points to your look (out of 10). Don't try to pick your own style, discuss it with the barber, and those four weeks are pretty brutal, you won't like it, but there's no way to know what if any kind of facial hair you can have. Not shaving every morning takes an additional -3 minutes of your life, and this time both in wall time and projects. Going to the barber takes 30 minutes, so the net time commitment is -3 minutes times 21 days = -63 minutes and 30 minutes to find and go to a barber for their styling shave. so this, too, takes a negative amount of time.

the only people who aren't able to accept negative time for free are those who have passed away and are no longer with us. live while you can.

Re: Let artificial intelligence guess your attractiveness and age

#94
post #4

Obligatory..."what's new?" Especially now that the site's down. I mean, is it different from Project Oxford, the Microsoft API that's been around for awhile and is still quite amazing? https://www.projectoxford.ai/demo/vision#Analysis I actually tried it out early this morning, to compare it with a stock install of OpenCV 3. It got the faces correct, and the ages very well too. Here are its guesses for the Star Wars…

the flag for "Adult Content" reminds me of anecdotes about mechanical turk workers standing in to keep uploads child friendly, now mixed with Age Identification this gets a new perspective. Is this in active use somewhere?

Re: Let artificial intelligence guess your attractiveness and age

#95

Very brave of you to put the founder photos on the front page...

You've got to admire their honesty for not putting an image matching algorithm in to automatically say the founders are hot! https://goo.gl/photos/jv82LHNQKxrt1Ce88

Re: Let artificial intelligence guess your attractiveness and age

#97
post #14

Hi, I'm Rasmus and worked on the algorithms behind faces.ethz.ch. If you have any technical questions, let me know! Sorry for having some stability issues, we got much more traffic than expected and are currently working hard to fix everything!

Hi Rasmus,

Cool work! I'm curious regarding the training dataset. What is the distribution of faces by race/age? Also regarding the raters, what is their distribution? (Race/age/cultural background)?

It's widely known that attractiveness is heavily dependent on cultural upbringing.

Signed, Butthurt dev whose best pic only rated an "OK".

EDIT: You also rated Yoona (a Korean pop star) as just "Nice": http://imgur.com/uJVnQ9S. I guess that makes me feel better about my "OK". I'd stay out of Korea if I were you---I hear their fans aren't very forgiving.

Re: Let artificial intelligence guess your attractiveness and age

#99
post #15
post #5

Startup idea #384826 - "Does this outfit make me look fat?" As A Service

I actually thought an interesting idea is an app that recommends you which dating site to use based on your face. If you're good looking you can pretty much use whatever. But for someone like myself who is ethnic and not visually attractive, my success rate is really low on certain sites and acceptable on certain apps. For example, my performance on Match (Graphic I made: http://i.imgur.com/UZuSzD9.png ) was pretty w…

When I was aggressively dating, I found that my success was prone to incredible fluctuations that I could never pin down. Keeping the site constant, anyway. I do suffer somewhat for being a certain flavor of queer that excludes a lot of heteronormative people.

Re: Let artificial intelligence guess your attractiveness and age

#100

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.

Get rid of the headphones and try again

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!

"Interestingly, 44.81% of the male ratings are positive, while only 8.58% of the female ratings are positive. Due to this strong bias of ratings by women, we only predict the ratings of men."

So, the algorithm learns to rate me as a heterosexual man?

From: http://arxiv.org/pdf/1510.07867v1.pdf

Post reply on HN