Deep fake detection using a sparse auto encoder with a graph capsule dual graph CNN

Abstract

Deepfake (DF) is a kind of forged image or video that is developed to spread misinformation and facilitate vulnerabilities to privacy hacking and truth masking with advanced technologies, including deep learning and artificial intelligence with trained algorithms. This kind of multimedia manipulation, such as changing facial expressions or speech, can be used for a variety of purposes to spread misinformation or exploitation. This kind of multimedia manipulation, such as changing facial expressions or speech, can be used for a variety of purposes to spread misinformation or exploitation. With the recent advancement of generative adversarial networks (GANs) in deep learning models, DF has become an essential part of social media. To detect forged video and images, numerous methods have been developed, and those methods are focused on a particular domain and obsolete in the case of new attacks/threats. Hence, a novel method needs to be developed to tackle new attacks. The method introduced in this article can detect various types of spoofs of images and videos that are computationally generated using deep learning models, such as variants of long short-term memory and convolutional neural networks. The first phase of this proposed work extracts the feature frames from the forged video/image using a sparse autoencoder with a graph long short-term memory (SAE-GLSTM) method at training time. The first phase of this proposed work extracts the feature frames from the forged video/image using a sparse autoencoder with a graph long short-term memory (SAE-GLSTM) method at training time. The proposed DF detection model is tested using the FFHQ database, 100K-Faces, Celeb-DF (V2) and WildDeepfake. The evaluated results show the effectiveness of the proposed method.

A review of deep learning‐based approaches for deepfake content detection

“…T A B L E 2 Summarized works considering CNN sorted by year and alphabetical order. Along the same lines, the work carried out by (Venkatachalam et al, 2022) proposed a two-level deepfake detection in which the first phase concerns the task of extracting feature frames from the forged image using a sparse autoencoder enhanced by a graph long-short term memory and in the second phase fed the extracted features as input to a capsule network. Experiments were conducted using Flickr-Faces-HQ (FFHQ), 100 K-Faces, Celeb-DF, and WildDeepfake datasets demonstrating good generalization and effectiveness in detecting deepfake images.…”

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

“…R. Saravana Ram, M. Vinoth Kumar, Tareq M. Al-shami, Mehedi Masud, Hanan Aljuaid and Mohamed Abouhawwash in [5] suggested extracting features from the input deepfake image using fuzzy clustering. Kandasamy V, Hubálovsk, and Trojovsk [14] revealed the Deep learning approach with two levels for detecting deepfake photos and videos. To extract features from face images, the recommended SAE technique is employed.…”

A Review of Deep Learning-based Approaches for Deepfake Content Detection

“…Along the same lines, the work carried out by Venkatachalam et al [26] proposed a two-level deepfake detection in which the first phase concerns the task of extracting feature frames from the forged image using a sparse autoencoder enhanced by a graph long-short term memory and in the second phase fed the extracted features as input to a capsule network. Experiments were conducted using Flickr-Faces-HQ (FFHQ), 100K-Faces, Celeb-DF, and WildDeepfake datasets demonstrating good generalization and effectiveness in detecting deepfake images.…”