Computer scientists at the “University of Waterloo” have developed a smartphone app that helps its users learn the “art” of making very good “selfies” photos.
The application has an algorithm that guides the user to put the camera in the right spot, so that it can capture the best possible picture.
“Selfies have evolved into a normal way for people to express themselves and their experiences; but not all of the selfies are the same”,
says Dan Vogel, a professor of computer science at the university.
“Unlike other applications that enhance a photo after you shoot it, this system provides guidance, which means that the user is actually learning why his photo will be better”.
By developing the algorithm, Vogel and Kean Lee, a former postgraduate student, originally took “3D” digital scans of people of “mediocre” appearance. Then they took hundreds of “virtual” selfies, writing code to control a smartphone virtual camera and computer-generated lighting, which allowed them to explore different synthesis principles, including the direction of lighting, the position and the size of the face, etc.
Then, using an online crowdsourcing service, they put thousands of people to vote for the best virtual “selfies”, and created mathematical models on the patterns of votes, to develop an algorithm that guides the users to pull the best “selfie”.
Later, they put “normal” people to take “selfies” with a conventional camera application and the camera application using the algorithm. Based on online ratings, there was an improvement of 26% on “selfies” taken with the application of “University of Waterloo” researchers.
“It’s just simply the beginning”,
says Vogel.
“We can expand the variables to include other factors, such as hair style, smile, and even clothes”
notes emphatically.
Source: www.naftemporiki.gr
(Συνολικές Επισκέψεις: / Total Visits: 12)
(Σημερινές Επισκέψεις: / Today's Visits: 1)