Nvidia’s scary AI generates humans that look 100 % real!

on . Posted in Articles of Interest

SANTA CLARA, Kalifornia (PNN) - December 17, 2018 - Believe it or not, all these faces are fake. They have been synthesized by Nvidia’s new AI algorithm, a generative adversarial network capable of automagically creating humans, cats, and even cars.

The technology works so well that we can expect synthetic image search engines soon - just like Google’s, but generating new fake images on the fly that look real. Yes, you know where that is going - and sure, it can be a lot of fun, but also scary.

According to Nvidia, its GAN is built around a concept called “style transfer”. Rather than trying to copy and paste elements of different faces into a frankenperson, the system analyzes three basic styles - coarse, middle, and fine - and merges them transparently into something completely new.

Coarse styles include parameters such as pose, the face’s shape, or the hairstyle. Middle styles include facial features, like the shape of the nose, cheeks, or mouth. Finally, fine styles affect the color of the face’s features like skin and hair.

According to the scientists, the generator is “capable of separating inconsequential variation from high-level attributes” too, in order to eliminate noise that is irrelevant for the new synthetic face.

For example, it can distinguish a hairdo from the actual hair, eliminating the former while applying the latter to the final photo. It can also specify the strength of how styles are applied to obtain more or less subtle effects.

Not only is the generative adversarial network capable of autonomously creating human faces, but it can do the same with animals like cats. It can even create new cars and even bedrooms.

Nvidia’s system is not only capable of generating completely new synthetic faces, but it can also seamlessly modify specific features of real people, like age, hair or skin color.

The applications for such a system are amazing. From paradigm-changing synthetic free-to-use image search pages that may be the end of stock photo services to people accurately previewing hair styling changes; and of course, porn.

However, the potential for fraud is immeasurable, and the possibility of corrupt governments using this technology to further encroach on the freedom and individuality of people cannot be dismissed.

Some technology must be devised to counteract this new threat.

Eulogies

Eulogy for an Angel
1992-Dec. 20, 2005

My Father
1918-2010

brents dad

Dr. Stan Dale
1929-2007

stan dale

A. Solzhenitsyn
1918-2008

solzhenitsyn

Patrick McGoohan
1928-2009

mcgoohan

Joseph A. Stack
1956-2010

Bill Walsh
1931-2007

Walter Cronkite
1916-2009

Eustace Mullins
1923-2010

Paul Harvey
1918-2009

Don Harkins
1963-2009

Joan Veon
1949-2010

David Nolan
1943-2010

Derry Brownfield
1932-2011

Leroy Schweitzer
1938-2011

Vaclav Havel
1936-2011

Andrew Breitbart
1969-2012

Dick Clark
1929-2012

Bob Chapman
1935-2012

Ray Bradbury
1920-2012

Tommy Cryer
1949-2012

Andy Griffith
1926-2012

Phyllis Diller
1917-2012

Larry Dever
1926-2012

Brian J. Chapman
1975-2012

Annette Funnicello
1942-2012

Margaret Thatcher
1925-2012

Richie Havens
1941-2013

Jack McLamb
1944-2014

James Traficant
1941-2014

jim traficant

Dr. Stan Monteith
1929-2014

stan montieth

Leonard Nimoy
1931-2015

Leonard Nimoy

Stan Solomon
1944-2015

Stan Solomon

B. B. King
1926-2015

BB King

Irwin Schiff
1928-2015

Irwin Schiff

DAVID BOWIE
1947-2016

David Bowie

Muhammad Ali
1942-2016

Muhammed Ali

GENE WILDER
1933-2016

gene wilder

phyllis schlafly
1924-2016

phylis schafly

John Glenn
1921-2016

John Glenn

Charles Weisman
1954-2016

Charles Weisman

Carrie Fisher
1956-2016

Carrie Fisher

Debbie Reynolds
1932-2016

Debbie Reynolds

Roger Moore
1917-2017

Roger Moore

Adam West
1928-2017

Adam West

JERRY LEWIS
1926-2017

jerry lewis

HUGH HEFNER
1926-2017

Hugh Hefner

PROF. STEPHEN HAWKING
1942-2018

Hugh Hefner 

ART BELL
1945-2018

Art Bell

DWIGHT CLARK
1947-2018

dwight clark

CARL MILLER
1952-2017

Carl Miller

HARLAN ELLISON
1934-2018

Harlan Ellison

STAN LEE
1922-2018

stan lee