What question did this study set out to answer?

To develop a forensic framework that detects deepfakes with identity-awareness using biometric embeddings and CNNs.

February 24, 2026Open Access

Implementation and Performance Analysis of an Identity-Aware Deepfake Forensic Framework Using 128-Dimensional Biometric Embeddings and CNN-Based Verification

Key Points

To develop a forensic framework that detects deepfakes with identity-awareness using biometric embeddings and CNNs.
Integrated 128-dimensional biometric embedding retrieval
Utilized vector similarity search for efficient matching
Implemented a Retrieve-Verify architecture
Employed FAISS indexing for biometric retrieval
Applied CNN for authenticity classification
Achieved 83.6% accuracy on a dataset of 5,000 images
Enabled identification of potential victims in manipulated media
Transformed the approach from general detection to forensic attribution

Abstract

Existing deepfake detection systems lack identity-awareness and cannot determine the potential victim associated with manipulated media. This paper presents an identity-aware forensic framework that integrates 128-dimensional facial embedding retrieval using vector similarity search with CNN-based verification. Following a Retrieve-Verify architecture, the system uses FAISS indexing for biometric retrieval followed by CNN authenticity classification. The system achieved a validated accuracy of 83.6% on a dataset of 5,000 images, transforming detection into a forensic attribution problem.

Implementation and Performance Analysis of an Identity-Aware Deepfake Forensic Framework Using 128-Dimensional Biometric Embeddings and CNN-Based Verification

Key Points

Abstract

Cite This Study