Gumbel Rao Monte Carlo based Bi-Modal Neural Architecture Search for Audio-Visual Deepfake Detection

Deepfakes pose a critical threat to biometric authentication systems by generating highly realistic synthetic media. Existing multimodal deepfake detectors often struggle to adapt to diverse data and rely on simple fusion methods. To address these challenges, we propose Gumbel-Rao Monte Carlo Bi-mod...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Aravinda Reddy PN, Ramachandra, Raghavendra, Krothapalli Sreenivasa Rao, Pabitra Mitra Vinod Rathod
Format Paper
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 09.10.2024
Subjects
Audio data
Deception
Deepfake
Parameter identification
Searching
Online AccessGet full text

Cover

More Information
Summary:Deepfakes pose a critical threat to biometric authentication systems by generating highly realistic synthetic media. Existing multimodal deepfake detectors often struggle to adapt to diverse data and rely on simple fusion methods. To address these challenges, we propose Gumbel-Rao Monte Carlo Bi-modal Neural Architecture Search (GRMC-BMNAS), a novel architecture search framework that employs Gumbel-Rao Monte Carlo sampling to optimize multimodal fusion. It refines the Straight through Gumbel Softmax (STGS) method by reducing variance with Rao-Blackwellization, stabilizing network training. Using a two-level search approach, the framework optimizes the network architecture, parameters, and performance. Crucial features are efficiently identified from backbone networks, while within the cell structure, a weighted fusion operation integrates information from various sources. By varying parameters such as temperature and number of Monte carlo samples yields an architecture that maximizes classification performance and better generalisation capability. Experimental results on the FakeAVCeleb and SWAN-DF datasets demonstrate an impressive AUC percentage of 95.4\%, achieved with minimal model parameters.
ISSN:2331-8422