AFMB-Net

Abstract

With advances in deepfake generating technology, it is getting increasingly difficult to detect deepfakes. Deepfakes can be used for many malpractices such as blackmail, politics, social media, etc. These can lead to widespread misinformation and can be harmful to an individual or an institution’s reputation. It has become important to be able to identify deepfakes effectively, while there exist many machine learning techniques to identify them, these methods are not able to cope up with the rapidly improving GAN technology which is used to generate deepfakes. Our project aims to identify deepfakes successfully using machine learning along with Heart Rate Analysis. The heart rate identified by our model is unique to each individual and cannot be spoofed or imitated by a GAN and is thus susceptible to improving GAN technology. To solve the deepfake detection problem we employ various machine learning models along with heart rate analysis to detect deepfakes.

Exploring Deepfake Detection: Techniques, Datasets and Challenges

“…The study by [44] combination of deep and traditional motion magnification is developed that achieves 97.17% accuracy. The study by [45] AFMB-Net employs ML and heart rate analysis, offering deepfake detection with unforgeable heart rate, advancing GAN technology, even in low-quality videos.…”

Developing an Interferogram-Based Module with Machine Learning for Maintaining Leveling of Glass Substrates

“…Interference patterns are of utmost importance in various domains, such as precision measurement and sensing, where the precise recognition of these fringes is crucial for obtaining dependable information. In addition to Inception v3, there are other modules available for image recognition, including Inception v4, Xception, Residual Net, and many others [20,21]. These modules have been extensively utilized and validated, each with its own unique architecture and performance advantages.…”

Artificial Intelligence and New Threat Vectors: Using Scenario Planning for Trend Forecasting

“…Notably, a method by A. Vinay et al combining heart-rate analysis with machine learning, leverages unique individual heart-rate patterns that GANs cannot mimic. 26 Similar approaches have been taken with DeepVision, which detects deepfakes by analyzing human eye-blinking patterns, 27 and DeFakePro, which employs Electrical Network Frequency signals to authenticate media broadcasts in online video conferencing tools. 28 Another recent strategy recommends using a specific generative model to create a deepfake, capitalizing on the residuals of the generator learned by convolutional neural network-based deepfake detection methods.…”