Deepfake Video Detection Using Generative Convolutional Vision Transformer

Deepfakes have raised significant concerns due to their potential to spread false information and compromise digital media integrity. In this work, we propose a Generative Convolutional Vision Transformer (GenConViT) for deepfake video detection. Our model combines ConvNeXt and Swin Transformer models for feature extraction, and it utilizes Autoencoder and Variational Autoencoder to learn from the latent data distribution. By learning from the visual artifacts and latent data distribution, GenConViT achieves improved performance in detecting a wide range of deepfake videos. The model is trained and evaluated on DFDC, FF++, DeepfakeTIMIT, and Celeb-DF v2 datasets, achieving high classification accuracy, F1 scores, and AUC values. The proposed GenConViT model demonstrates robust performance in deepfake video detection, with an average accuracy of 95.8% and an AUC value of 99.3% across the tested datasets. Our proposed model addresses the challenge of generalizability in deepfake detection by leveraging visual and latent features and providing an effective solution for identifying a wide range of fake videos while preserving media integrity. The code for GenConViT is available at https://github.com/erprogs/GenConViT.

Longitudinal Analysis of Mask and No-Mask on Child Face Recognition

Face is one of the most widely employed traits for person recognition, even in many large-scale applications. Despite technological advancements in face recognition systems, they still face obstacles caused by pose, expression, occlusion, and aging variations. Owing to the COVID-19 pandemic, contactless identity verification has become exceedingly vital. To constrain the pandemic, people have started using face mask. Recently, few studies have been conducted on the effect of face mask on adult face recognition systems. However, the impact of aging with face mask on child subject recognition has not been adequately explored. Thus, the main objective of this study is analyzing the child longitudinal impact together with face mask and other covariates on face recognition systems. Specifically, we performed a comparative investigation of three top performing publicly available face matchers and a post-COVID-19 commercial-off-the-shelf (COTS) system under child cross-age verification and identification settings using our generated synthetic mask and no-mask samples. Furthermore, we investigated the longitudinal consequence of eyeglasses with mask and no-mask. The study exploited no-mask longitudinal child face dataset (i.e., extended Indian Child Longitudinal Face Dataset) that contains $26,258$ face images of $7,473$ subjects in the age group of $[2, 18]$ over an average time span of $3.35$ years. Due to the combined effects of face mask and face aging, the FaceNet, PFE, ArcFace, and COTS face verification system accuracies decrease approximately $25\%$, $22\%$, $18\%$, $12\%$, respectively.

Determining Sequence of Image Processing Technique (IPT) to Detect Adversarial Attacks

Developing secure machine learning models from adversarial examples is challenging as various methods are continually being developed to generate adversarial attacks. In this work, we propose an evolutionary approach to automatically determine Image Processing Techniques Sequence (IPTS) for detecting malicious inputs. Accordingly, we first used a diverse set of attack methods including adaptive attack methods (on our defense) to generate adversarial samples from the clean dataset. A detection framework based on a genetic algorithm (GA) is developed to find the optimal IPTS, where the optimality is estimated by different fitness measures such as Euclidean distance, entropy loss, average histogram, local binary pattern and loss functions. The "image difference" between the original and processed images is used to extract the features, which are then fed to a classification scheme in order to determine whether the input sample is adversarial or clean. This paper described our methodology and performed experiments using multiple data-sets tested with several adversarial attacks. For each attack-type and dataset, it generates unique IPTS. A set of IPTS selected dynamically in testing time which works as a filter for the adversarial attack. Our empirical experiments exhibited promising results indicating the approach can efficiently be used as processing for any AI model.

Generalizable Adversarial Examples Detection Based on Bi-model Decision Mismatch

Deep neural networks (DNNs) have shown phenomenal success in a wide range of applications. However, recent studies have discovered that they are vulnerable to Adversarial Examples, i.e., original samples with added subtle perturbations. Such perturbations are often too small and imperceptible to humans, yet they can easily fool the neural networks. Few defense techniques against adversarial examples have been proposed, but they require modifying the target model or prior knowledge of adversarial examples generation methods. Likewise, their performance remarkably drops upon encountering adversarial example types not used during the training stage. In this paper, we propose a new framework that can be used to enhance DNNs' robustness by detecting adversarial examples. In particular, we employ the decision layer of independently trained models as features for posterior detection. The proposed framework doesn't require any prior knowledge of adversarial examples generation techniques, and can be directly augmented with unmodified off-the-shelf models. Experiments on the standard MNIST and CIFAR10 datasets show that it generalizes well across not only different adversarial examples generation methods but also various additive perturbations. Specifically, distinct binary classifiers trained on top of our proposed features can achieve a high detection rate (>90%) in a set of white-box attacks and maintain this performance when tested against unseen attacks.