![Social Networks Social Networks](http://a.fsdn.com/sd/topics/social_64.png)
![Software Software](http://a.fsdn.com/sd/topics/software_64.png)
![The Internet The Internet](http://a.fsdn.com/sd/topics/internet_64.png)
Deep-Live-Cam Goes Viral, Allowing Anyone To Become a Digital Doppelganger (arstechnica.com) 17
An anonymous reader quotes a report from Ars Technica: Over the past few days, a software package called Deep-Live-Cam has been going viral on social media because it can take the face of a person extracted from a single photo and apply it to a live webcam video source while following pose, lighting, and expressions performed by the person on the webcam. While the results aren't perfect, the software shows how quickly the tech is developing -- and how the capability to deceive others remotely is getting dramatically easier over time. The Deep-Live-Cam software project has been in the works since late last year, but example videos that show a person imitating Elon Musk and Republican Vice Presidential candidate J.D. Vance (among others) in real time have been making the rounds online. The avalanche of attention briefly made the open source project leap to No. 1 on GitHub's trending repositories list (it's currently at No. 4 as of this writing), where it is available for download for free. [...]
Like many open source GitHub projects, Deep-Live-Cam wraps together several existing software packages under a new interface (and is itself a fork of an earlier project called "roop"). It first detects faces in both the source and target images (such as a frame of live video). It then uses a pre-trained AI model called "inswapper" to perform the actual face swap and another model called GFPGAN to improve the quality of the swapped faces by enhancing details and correcting artifacts that occur during the face-swapping process. The inswapper model, developed by a project called InsightFace, can guess what a person (in a provided photo) might look like using different expressions and from different angles because it was trained on a vast dataset containing millions of facial images of thousands of individuals captured from various angles, under different lighting conditions, and with diverse expressions.
During training, the neural network underlying the inswapper model developed an "understanding" of facial structures and their dynamics under various conditions, including learning the ability to infer the three-dimensional structure of a face from a two-dimensional image. It also became capable of separating identity-specific features, which remain constant across different images of the same person, from pose-specific features that change with angle and expression. This separation allows the model to generate new face images that combine the identity of one face with the pose, expression, and lighting of another.
Like many open source GitHub projects, Deep-Live-Cam wraps together several existing software packages under a new interface (and is itself a fork of an earlier project called "roop"). It first detects faces in both the source and target images (such as a frame of live video). It then uses a pre-trained AI model called "inswapper" to perform the actual face swap and another model called GFPGAN to improve the quality of the swapped faces by enhancing details and correcting artifacts that occur during the face-swapping process. The inswapper model, developed by a project called InsightFace, can guess what a person (in a provided photo) might look like using different expressions and from different angles because it was trained on a vast dataset containing millions of facial images of thousands of individuals captured from various angles, under different lighting conditions, and with diverse expressions.
During training, the neural network underlying the inswapper model developed an "understanding" of facial structures and their dynamics under various conditions, including learning the ability to infer the three-dimensional structure of a face from a two-dimensional image. It also became capable of separating identity-specific features, which remain constant across different images of the same person, from pose-specific features that change with angle and expression. This separation allows the model to generate new face images that combine the identity of one face with the pose, expression, and lighting of another.
Stop The World (Score:4, Funny)
I want to get off!
Re:Stop The World (Score:5, Funny)
I want to get off!
There are other types of deep live cams for that, I'm sure.
Re: (Score:2)
Re: (Score:2)
"I don't want to live on this planet anymore." --Professor Farnsworth
ID.ME (Score:5, Insightful)
The ID.ME login junk that that IRS and Social Security Administration are pushing (now requiring, I believe) to communicate with them has video chat as the fallback verification option for when they can't verify you by asking questions based on the Equifax data ("Have you ever applied for a credit card with: KOHLS STORES LLC")
They were mistaken if they thought getting a video stream from someone was a reliable way to verify identity.
But yeah, with this live deepfake software out there, they'll have to rethink the entire procedure when it's completely cracked within a couple years.
Re: (Score:2)
I think something like Apple Face ID already goes a step further and makes use of depth information from the cameras. So possibly they could incorporate that type of information which would be more difficult to spoof from a 2d image.
Re: (Score:2)
The "you must have a webcam" requirement seemed a little ridiculous on its own for something like filing taxes. Bumping that up to "You must have the new iPhone" will make it completely useless as a fallback method for the majority of people.
Only downsides, no upside (Score:1)
Plenty of potential upsides (Score:4, Insightful)
Secondly, everyone having access to a convincingly fake appearance online should eventually destroy the influencer market, while redefining perceptions of social media in favour of more traditional communication methods. No amount of online ID checks will be able to combat the resharing of nonsense, which is where public social media websites fall down, but private instant messaging between real world friends does not.
Finally, people might just leave the house more. Since chasing clout online will mean far less, people might actually do more things for their local community to gain recognition instead. The value of contributing to the benefit of ones own local area was lost when people realised they could derive far more value from doing things online, that could change!
like nuclear weapons for children (Score:2)
especially with the elections. omg..
There aren't any applications of this that essentially aren't fraud.
It's made specifically to do Fraud. Get it now?
Jeeezuzzzz.. Who wouldn't want that?
Re: (Score:2)
George Clooney (Score:2)
A variety of oddly youthful celebrities will be attending the corporate telecons in the near future. So bright I gotta wear shades etc.