Real-time deepfake videos, heralded as the bringers of a new age of internet uncertainty, appear to have a fundamental flaw: They can't handle side profiles. That's the conclusion drawn in a report [PDF] from Metaphysic.ai, which specializes in 3D avatars, deepfake technology and rendering 3D images from 2D photographs. In tests it conducted using popular real-time deepfake app DeepFaceLive, a hard turn to the side made it readily apparent that the person on screen wasn't who they appeared to be. Multiple models were used in the test - several from deepfake communities and models included in DeepFaceLive - but a a 90-degree view of the face caused flickering and distortion as the Facial Alignment Network used to estimate poses struggled to figure out what it was seeing. "Most 2D-based facial alignments algorithms assign only 50-60 percent of the number of landmarks from a front-on face view to a profile view," said Metaphysic.ai contributor Martin Anderson, who wrote the study's blog post.

There is an interesting vulnerability in video deepfakes that, to date, has been generally overlooked by the security research community, perhaps because'live', real-time deepfakes in video calls have not been a major cause for concern until very recently. For a number of reasons, which we'll examine in this article, deepfakes are not usually very good at recreating profile views: The above examples are taken* from a session with tech exponent and commenter Bob Doyle, who agreed to run some tests with us, using DeepFaceLive to change his appearance to that of a series of popular celebrities. DeepFaceLive is a live-streaming version of the popular DeepFaceLab software, and is capable of creating alternate video identities in real-time. For these tests, Bob used a mix of models downloaded from deepfake communities, as well as some community-trained models that come bundled with the DeepFaceLive application. From more or less face-on viewpoints, most of the celebrity recreations are quite effective, and some are very convincing even at fairly acute angles – until the facial angle hits a full 90 .

You're more likely to have your mother's cheekbones than her eyes, new research suggests. Researchers studied the facial features of 1,000 female twins to find parts of the face that are likely to be controlled by genetics. They used their results to create interactive face maps that reveal the features you're most likely to inherit from your parents. Biological traits such as facial features are influenced by genes and'environmental' factors including the socioeconomic conditions a person grew up in. Professor Giovanni Montana, from King's College London, said: 'The notion that our genes control our face is self-evident.

With an estimated 50 million users swiping on Tinder every month, it can be easy for your profile to be overlooked. But Tinder has introduced a new feature that could help you get noticed in a sea of profiles. Tinder Boost is a new premium feature that places paid users at the top for 30 minutes, giving them up to 10 times more profile views than the standard method. After being announced earlier this month, the update is now being rolled out for Tinder users worldwide over the next few days. Tinder will ask users if they'd like to activate Boost while swiping through profiles in the app.