AI-Powered Deepfake Detection Using CNN and Vision Transformer Architectures

Sifatullah Sheikh Urmi, Kirtonia Nuzath Tabassum Arthi, Md Al-Imran
1/3/2026
cs.CVcs.AIcs.LG

No official code found. Best implementation: SifatSwapnil2022/conf_AI-Powered_Deepfake_Detection_codes (0 stars, last commit 2026-01-03).

Abstract

The increasing use of artificial intelligence generated deepfakes creates major challenges in maintaining digital authenticity. Four AI-based models, consisting of three CNNs and one Vision Transformer, were evaluated using large face image datasets. Data preprocessing and augmentation techniques improved model performance across different scenarios. VFDNET demonstrated superior accuracy with MobileNetV3, showing efficient performance, thereby demonstrating AI's capabilities for dependable deepfake detection.

View on arXivView PDF

Code Implementations(1)

SifatSwapnil2022/conf_AI-Powered_Deepfake_Detection_codesCommunity73%

AI-Powered Deepfake Detection Using CNN and Vision Transformer Architectures

01Jupyter NotebookDec 19, 20259 months ago

Similar Papers

Deepfake Detection with Deep Learning: Convolutional Neural Networks versus Transformers

90% match2023

Classifying Deepfakes Using Swin Transformers

88% match2025

State-of-the-art AI-based Learning Approaches for Deepfake Generation and Detection, Analyzing Opportunities, Threading through Pros, Cons, and Future Prospects

88% match2025

Deepfake Detection Via Facial Feature Extraction and Modeling

88% match1 repo12025

Towards Trustworthy AI: Secure Deepfake Detection using CNNs and Zero-Knowledge Proofs

88% match2025

Evaluating Deepfake Detectors in the Wild

88% match1 repo62025

Cite this paper

@article{urmi2026aipowered,
  title  = {AI-Powered Deepfake Detection Using CNN and Vision Transformer Architectures},
  author = {Sifatullah Sheikh Urmi and Kirtonia Nuzath Tabassum Arthi and Md Al-Imran},
  year   = {2026},
  eprint = {2601.01281},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2601.01281v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2601.01281.svg)](https://codeofpaper.com/paper/2601.01281)

Discussion