A Quantum-Hybrid Framework for Enhanced Deepfake Detection: Integrating QAOA-Based Feature Selection With Quantum-Inspired Attention Mechanisms

Deepfake technology has created major new challenges to the authentication of digital media and the security of digital information. The traditional methods using deep learning to detect deepfakes have shown good results. However, many of these traditional methods suffer from redundant features, imbalanced classes, and poor interpretability. This paper introduces a new hybrid quantum-classical framework for detecting deepfakes that combines InceptionResnetV1 to extract high dimensional representation with a Convolutional Neural Network (CNN), and a Quantum Inspired Attention Mechanism to select the most important features to process. Additionally, this paper proposes a Quantum Approximate Optimization Algorithm (QAOA) for feature selection based on high dimensional representations of images obtained from an InceptionResnetV1. QAOA selects the most informative features to be input into the CNN and helps reduce dimensionality while maintaining the ability to discriminate between classes of data. The selected features are then passed to a Multi-Head Quantum Inspired Attention Layer that utilizes quantum interference patterns to re-weight feature importance dynamically. The use of a focal Loss function with label smoothing and confidence penalty is also proposed to balance the class distribution in the deepfake datasets. Experiments conducted on the FaceForensics++ benchmark dataset show that the quantum-hybrid method outperforms classical methods in terms of accuracy, precision, recall, and calibration metrics. Grad-CAM is used as a visualization tool to provide interpretable output by showing which facial areas the model is focusing on when identifying forensic evidence.

• Publication year

  2026

• Venue

  IEEE Access

• Publication date

  Unknown publication date

• Fields of study

  Physics, Computer Science

• Identifiers
  DOI 10.1109/ACCESS.2026.3659021
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Video deepfake detection using a hybrid CNN-LSTM-Transformer model for identity verification

  2025cited by this paper
• Explainable Deep Learning Framework for Binary Corrosion Image Classification Using Grad-CAM

  2025cited by this paper
• Combining MTCNN and Enhanced FaceNet with Adaptive Feature Fusion for Robust Face Recognition

  2025cited by this paper
• FAME: A Lightweight Spatio-Temporal Network for Model Attribution of Face-Swap Deepfakes

  2025cited by this paper
• Automatic face recognition based on enhanced Vggface- 16 model in an unconstrained environment using transfer learning

  2025cited by this paper
• Deepfake video detection using CNN and RNN with OPTICAL FLOW features

  2024cited by this paper
• PQML: Enabling the Predictive Reproducibility on NISQ Machines for Quantum ML Applications

  2024cited by this paper
• Multiscale Transformer and Attention Mechanism for Magnetic Spatiotemporal Sequence Localization

  2024cited by this paper
• InceptionNeXt: When Inception Meets ConvNeXt

  2023cited by this paper
• Evidence of scaling advantage for the quantum approximate optimization algorithm on a classically intractable problem

  2023cited by this paper
• Efficient balanced focal loss function for manipulated images detection

  2021cited by this paper
• Improved Residual Networks for Image and Video Recognition

  2020cited by this paper
• Deepfake Video Detection through Optical Flow Based CNN

  2019cited by this paper
• EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks

  2019cited by this paper
• FaceForensics++: Learning to Detect Manipulated Facial Images

  2019influential reference
• Exposing Deep Fakes Using Inconsistent Head Poses

  2018cited by this paper
• Capsule-forensics: Using Capsule Networks to Detect Forged Images and Videos

  2018cited by this paper
• VGGFace2: A Dataset for Recognising Faces across Pose and Age

  2017cited by this paper
• Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning

  2016cited by this paper
• A Quantum Approximate Optimization Algorithm

  2014cited by this paper

• No citing papers are available for this paper.