In this section we review the origin, malicious usage, and protection against deepfake technology. You will know how criminals use deep learning to attack the likes of IoT systems, spread fake news, and spoof various biometric systems including facial recognition. We’ll also provide some insight on deepfake detection architectures and standardization.
Stories
Writers
The first technology similar to facial deepfakes appeared in 1997 when the Video Rewrite tool was presented. It was based on automatic phoneme labeling that allowed matching an already existing footage to a new soundtrack. The tool was successfully applied to alter a few bits
8 min read
8 min read
Deepfake media have shown an alarming increase in recent times. According to expert reports, the amount of that type of media, including facial deepfakes, doubles every six months, as the tools and means to produce such fabricated media are becoming greatly available to the
12 min read
12 min read
Deepfake detection technology allows prompt recognition of a piece of fabricated media with high accuracy based on liveness signals. Distinguishing fake footage, photo or audio from authentic media is also necessary, as it helps to tackle the so-called liar’s dividend
9 min read
9 min read
Cheapfake — ( a combination of cheap and deepfake) — are a class of fake media that is easy and quick to produce, especially for amateur users. Compared to the traditional deepfakes, cheapfakes do not require highly specific skills such as coding, manual neural network
13 min read
13 min read
Deepfake dataset is a collection of artificially synthesized media, which can include photo, video, and audio materials designed in accordance with the deepfake standardization. Their origin was urged by the proliferation of deepfake media that represents a steadily growing social threat
9 min read
9 min read
Deepfake detection is a capability of recognizing falsified media and distinguishing it from bona fide visual or auditory data. Deepfake technology was invented in the late 1990s, but became a widespread phenomenon in 2017. As a result, researchers have voiced concerns about the
13 min read
13 min read
Deepfakes are a type of falsified media — audio or visual — produced with deep learning. They employ various techniques — such as face swapping or voice conversion — to mimic a target person’s appearance or voice. The technology can be traced back to 1997 when the first
10 min read
10 min read
Convolutional neural networks (CNNs) were first introduced in the 1980s. One of the first known examples of CNN was the Time Delay Neural Network (TDNN) developed in 1987 by Alex Waibel and his research team. TDNN was focused on speech recognition, including shift-invariant
9 min read
9 min read
Editors:
Olga KokoulinaDigital face manipulation has recently emerged as a significant threat to biometric systems. Although manipulation of images/photographs — photoshopping — has been a popular practice for many years, video manipulation has been relatively unknown. Video
Deepfake (derived from ‘deep learning’ and ‘fake’) is a falsified synthetic media — video, photo, or audio — which presents a certain action that was not performed by a given person in reality. This technology uses techniques of deep machine learning and Artificial Intelligence (AI) to
11 min read
11 min read
Everything you wanted to know about Liveness Detection, Spoofing Attacks and Antispoofing Measures.
Biometrics Security, Access Control, Spoofing Attacks, Liveness Check, Recognition Security System...
10
Stories
1
Writers
Contribute to Antispoofing Wiki
by Becoming an Author or CommenterIf you have a good knowledge of biometric security, you can take part in writing or editing articles for the Antispoofing Wiki and be among the authors of the project. Details can be found at the link below
If you have questions about the topics of articles or want to comment on them, you can do this through the light version of registration