Randomized Shame
Sub-Admin
Staff
- Joined
- Feb 18, 2023
- Posts
- 1,558
- Reputation
- 3,127
I found this article in a Discord server and I thought it was interesting.
"Using Artificial Intelligence to Quantify Sexual Dimorphism in Aesthetic Faces: Analysis of 100 Facial Points in 42 Caucasian Celebrities" is the name of the article and it was released in 2023, so I'm getting to it kind of late but lmao.
It's a simple read. I thought I'd see what others think about it though.
Their method for determining the ratios was actually interesting.
First, they said they used Apple's Vision Framework for the AI that determined the basic dots (or landmarks as they called it) on the face. They still had to manually input some landmarks themselves, though, so
Second, they determined each subject's white-to-white corneal diameter (the width of the colored part of the eye) in pixels, then they used that to determine the rest of the face. That's obviously an over simplification but I doubt anyone here actually gives a shit what the method was in detail.
It's more interesting than I'm making it out to be, but it might be something we could learn from. I don't know how people currently determine their ratios. The findings in general was just also interesting.
"Using Artificial Intelligence to Quantify Sexual Dimorphism in Aesthetic Faces: Analysis of 100 Facial Points in 42 Caucasian Celebrities" is the name of the article and it was released in 2023, so I'm getting to it kind of late but lmao.
It's a simple read. I thought I'd see what others think about it though.
Only caucasian celebrities aged 23 to 42 were included in this study.
The female celebrities and models included in their study were: Angelina Jolie, Alecia Beth Moore, Ashley Graham, Bella Hadid, Catherine Zeta-Jones, Christina Applegate, Cindy Crawford, Courtney Cox, Gigi Hadid, Gisele Bundchen, Gwyneth Paltrow, Jennifer Garner, Jodie Foster, Julia Roberts, Karlie Kloss, Kate Hudson, Kendall Jenner, Michelle Pfeiffer, Nicole Kidman, Rosie Huntington, and Sandra Bullock.
The male celebrities and models included in their study were: Adam Levine, Blake Shelton, Bradley Cooper, Channing Tatum, Chris Hemsworth, David Beckham, Leonardo DiCaprio, Mel Gibson, Arthur Kulkov, David Gandy, John Kortajarena, Sean O’Pry, Tobias Sorensen, Noah Mills, Ollie Edwards, Paul Rudd, Ryan Burns, Ryan Reynolds, Simon Nessman, Tom Cruise, and Tyson Ballou.
The female celebrities and models included in their study were: Angelina Jolie, Alecia Beth Moore, Ashley Graham, Bella Hadid, Catherine Zeta-Jones, Christina Applegate, Cindy Crawford, Courtney Cox, Gigi Hadid, Gisele Bundchen, Gwyneth Paltrow, Jennifer Garner, Jodie Foster, Julia Roberts, Karlie Kloss, Kate Hudson, Kendall Jenner, Michelle Pfeiffer, Nicole Kidman, Rosie Huntington, and Sandra Bullock.
The male celebrities and models included in their study were: Adam Levine, Blake Shelton, Bradley Cooper, Channing Tatum, Chris Hemsworth, David Beckham, Leonardo DiCaprio, Mel Gibson, Arthur Kulkov, David Gandy, John Kortajarena, Sean O’Pry, Tobias Sorensen, Noah Mills, Ollie Edwards, Paul Rudd, Ryan Burns, Ryan Reynolds, Simon Nessman, Tom Cruise, and Tyson Ballou.
Their method for determining the ratios was actually interesting.
First, they said they used Apple's Vision Framework for the AI that determined the basic dots (or landmarks as they called it) on the face. They still had to manually input some landmarks themselves, though, so
Second, they determined each subject's white-to-white corneal diameter (the width of the colored part of the eye) in pixels, then they used that to determine the rest of the face. That's obviously an over simplification but I doubt anyone here actually gives a shit what the method was in detail.
It's more interesting than I'm making it out to be, but it might be something we could learn from. I don't know how people currently determine their ratios. The findings in general was just also interesting.
