Telling if a person is homosexual or straight can be tricky and extremely hard. But did you know that a computer can detect a person’s sexual orientation just based on his or her face?
A new computer program can recognize whether a person is gay or straight by just looking at his or her facial features. It can identify this with uncanny accuracy of 91 percent!
The software uses photos of the faces of people from all walks of life.
A team of researchers at the Stanford University developed a software using an artificial intelligence algorithm that can identify whether a face belongs to a straight or gay person.
Published in the Journal of Personality and Social Psychology, the study shows that the software can identify the sexual orientation with 81 percent accuracy for males and 74 percent accuracy for females.
The computer can determine if a person is gay more accurately than humans.
Humans, on the other hand, are just able to determine a person’s sexual orientation through images with an accuracy of 62 percent for men and 54 percent for women.
When the computer was shown more photos of a person, the accuracy increased to 91 percent for men and 83 percent for women.
The researchers created the algorithm by extracting the facial features from more than 35,000 images from a dating website. The photos were from people between the ages of 18 and 40, with heterosexual and homosexual individuals equally represented.
So, how can the computer accurately determine who's gay and who's not by just looking at photos?
Well, the program uses fixed features such as the shape of the nose and grooming style. Moreover, gay men have longer noses, narrower jaws, and slimmer eyebrows. On the other hand, gay women tend to have larger jaws and smaller foreheads. They have more masculine features.
The other factors that were considered are grooming styles, the amount of makeup and the person’s smile.
The researchers, Yilun Wang and Michal Kosinski, concluded:
“Deep neural networks are more accurate than humans at detecting sexual orientation from facial images.”
Female Shoplifter Goes Wild, Gets Body Slammed After Being Caught
Frankly, she deserved it!
You'd think that getting caught shoplifting can be an embarrassing thing. But apparently, the female shoplifter in the video below had the guts to get mad and even assault the store employee who held her.
It's really bizarre but yes, everything's been caught on tape for the world to see. In the 4-minute video uploaded on YouTube by KodyXO, we see how the lady shoplifter went wild and crazy after she was captured by a Rite Aid employee for stealing two items at their store in Hillsboro, Oregon.
The Rite Aid worker grabbed her purse where she kept the stolen items.
Five Nurses Suspended After Admiring A Dead Patient’s Genitals
WTF! This is totally inappropriate!
Five nurses in the United States were in hot waters when they were heard making malicious comments about a dead patient's genitals. The nurses, who work at the Denver Health Medical Center in Colorado, received a three-week suspension.
The said incident reportedly happened between March 31 and April 3, during which the nurses inappropriately viewed a dead patient's body and talked about it. The incident remained unreported until May 8.
One of the nurses even opened the body bag and made sexual comments about the size of the patient's genitals.
Diver Becomes Like a Human Balloon After Surfacing From The Ocean Too Fast
This diver suffers from decompression sickness, or commonly called “the bends.”
Alejandro Ramos Martinez of Peru is an experienced diver who makes a living by descending to the depths of the ocean and going for some fresh catch. But one day, he made a mistake of surfacing rather too quickly, which left him looking like a balloon.
Martinez’s body swelled up during the diving accident from four years ago. As a result, he is now suffering from a condition called decompression sickness, also known as “the bends.” This happens when a diver rises from the depths too fast, causing dissolved nitrogen in the body to form large sacs that adhere to the muscles.