Researchers from the Max Planck Institute for Intelligent Systems
and the University of Texas at Dallas, and revealed new crowdshaping technology in the SIGGRAPH 2016 to create an accurate 3-D models of the body
using 2D images linguistic descriptions of mobilizing resources from the shape of the body.
stucture Talk system takes a single image and produce a 3-D body shapes that look like the person
and are accurate enough for the size of clothing.
They do so by using the help of 15 volunteers from the living body in the form of the image using 30 words or less. The researchers believe that this technology has applications in online shopping, games,
virtual reality and health care.
“Scanners, high-end and lower-cost alternatives have not yet become ubiquitous in consumer technology.
” said Dr. Michael J. Black,
managing director of perception at the Max Planck Institute for Intelligent Systems systems. “Capture 3-D shape of the stucture is still a challenge,
our goal is to make it easy and fun. Our research shows that people have a common understanding of the shape of the body,
which is expressed in our use of language.”
Machine learning system is used to detect this relationship
between us verbal descriptions of the bodies and the actual shape of their 3-D. Said Dr. Alice J. O’Toole, a professor at the School of Behavioral and Brain Sciences at the University of Texas at Dallas “description language provides rich features of the body due to the variable sex,
physical strength, attractiveness and health.”
This process of crowdshaping opens a practical way to examine the bodies without a scanner,” Dr. Stefan Streuber,
lead author of the study. In the paper, the modern body: crowdshaping embody realistic 3-D with the words, researchers found that a realistic 3-D bodies can be created using a few of the ten words.
They have demonstrated the application of “3-D photographers” that take a single picture of a celebrity.
The researchers also showed that they can extract measurements of bodies crowdshaped
with sufficient accuracy for many applications clothing sizing. So they create gods and 3-D characters from books using just a written description.
Japan tests innovative magnetic tether for slowing space junk
INDONESIA Earthquake:Number of homeless soars to 43,000
Jennifer Lawrence apologies for Hawaii ‘butt-scratching’ story
A History of PIA Accidents That Have Killed Over Many People!
South Korea lawmakers vote to impeach leader