SKDU at De-Factify 4.0: Vision Transformer with Data Augmentation for AI-Generated Image Detection

Shrikant Malviya, Neelanjan Bhowmik, Stamos Katsigiannis
3/24/2025
cs.CV

No official code found. Best implementation: skmalviya/ai_gen_image_defactify (2 stars, last commit 2025-03-02).

Abstract

The aim of this work is to explore the potential of pre-trained vision-language models, e.g. Vision Transformers (ViT), enhanced with advanced data augmentation strategies for the detection of AI-generated images. Our approach leverages a fine-tuned ViT model trained on the Defactify-4.0 dataset, which includes images generated by state-of-the-art models such as Stable Diffusion 2.1, Stable Diffusion XL, Stable Diffusion 3, DALL-E 3, and MidJourney. We employ perturbation techniques like flipping, rotation, Gaussian noise injection, and JPEG compression during training to improve model robustness and generalisation. The experimental results demonstrate that our ViT-based pipeline achieves state-of-the-art performance, significantly outperforming competing methods on both validation and test datasets.

View on arXivView PDF

Code Implementations(1)

skmalviya/ai_gen_image_defactifyCommunity100%

This repo is on the solution to Defactify 4.0 AI-Gen Image Detection Challenge (2024-25)

21PythonJan 3, 20251 years ago

Similar Papers

LoRA-TTT: Low-Rank Test-Time Training for Vision-Language Models

86% match2025

POINTS: Improving Your Vision-language Model with Affordable Strategies

86% match2024

Evidence Packing for Cross-Domain Image Deepfake Detection with LVLMs

86% match2026

ViTamin: Designing Scalable Vision Models in the Vision-Language Era

86% match1 repo2122026

Enhancing Vision-Language Model with Unmasked Token Alignment

85% match1 repo112024

LARE: Latent Augmentation using Regional Embedding with Vision-Language Model

85% match2024

Cite this paper

@article{malviya2025skdu,
  title  = {SKDU at De-Factify 4.0: Vision Transformer with Data Augmentation for AI-Generated Image Detection},
  author = {Shrikant Malviya and Neelanjan Bhowmik and Stamos Katsigiannis},
  year   = {2025},
  eprint = {2503.18812},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2503.18812v1}
}

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/2503.18812.svg)](https://codeofpaper.com/paper/2503.18812)

Discussion