Deep fake detection using cascaded deep sparse auto-encoder for effective feature selection

Abstract

In the recent research era, artificial intelligence techniques have been used for computer vision, big data analysis, and detection systems. The development of these advanced technologies has also increased security and privacy issues. One kind of this issue is Deepfakes which is the combined word of deep learning and fake. DeepFake refers to the formation of a fake image or video using artificial intelligence approaches which are created for political abuse, fake data transfer, and pornography. This paper has developed a Deepfake detection method by examining the computer vision features of the digital content. The computer vision features based on the frame change are extracted using a proposed deep learning model called the Cascaded Deep Sparse Auto Encoder (CDSAE) trained by temporal CNN. The detection process is performed using a Deep Neural Network (DNN) to classify the deep fake image/video from the real image/video. The proposed model is implemented using Face2Face, FaceSwap, and DFDC datasets which have secured an improved detection rate when compared to the traditional deep fake detection approaches.

Unveiling Spoofing Attempts: A DCGAN-based Approach to Enhance Face Spoof Detection in Biometric Authentication

“…The formula for calculating the Area Under the Curve (AUC) in a ROC curve is given in Eq. ( 7), 𝐴𝑈𝐶 = ∑𝑖 = 1𝑛 − 12(𝑥𝑖 + 1 − 𝑥𝑖) ⋅ (𝑦𝑖 + 𝑦𝑖 + 1) (7) where, (xi,yi) are the points of the ROC curve, and n is the total number of points. The AUC shows the integral of the ROC curve, which measures the overall evaluation of a binary classification approach.…”

“…A Deepfake detection technique utilizes computer vision characteristics extracted from digital context, employing the Cascaded Deep Sparse Auto Encoder (CDSAE) trained by temporal based CNN to analyze frame changes. Subsequently, a Deep Neural Network (DNN) is employed for classification, achieving enhanced accuracies of 98.7%, 98.5%, and 97.63% for the datasets like Face2Face, FaceSwap, and DFDC, through a feature selection approach [7]. Face swapping detection employed by deep transfer learning for, achieving true positive rates exceeding 96% with minimal false alarms, and providing uncertainty estimates for each prediction, crucial for system trust.…”

Combining Computer Vision Techniques and Intraframe Noise Methods to Detect a Deepfake

“…In order to create Deep Fake the Deep neural networks are used. Various recent technological advancements with deep learning techniques including auto encoders and GAN [6][7](Generative Adversarial Networks) are used to create fake faces which are applied mainly in the computer vision. Deep Fake detection method using the Haar wavelet transforms.…”