Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
AI GenDetect: A comprehensive approach to detecting AI-generated images
0
Zitationen
3
Autoren
2025
Jahr
Abstract
As AI-generated images get more and more sophisticated, people's concerns grow regarding the authenticity of visual content. This work proposes an approach called AI GenDetect, which detects AI-generated images using a curated dataset based on the combination of the CASIA dataset with real-world images. Using error level analysis as preprocessing and the power of CNNs, the model reached 91.83% accuracy in 9 epochs. Results include precision-recall metrics, a confusion matrix, and real-world testing scenarios, which show its practical effectiveness. AI GenDetect is a robust approach to image authentication, addressing the current challenges while paving the way for future advancements in safeguarding visual content.
Ähnliche Arbeiten
Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI
2019 · 8.339 Zit.
Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead
2019 · 8.211 Zit.
High-performance medicine: the convergence of human and artificial intelligence
2018 · 7.614 Zit.
Proceedings of the 19th International Joint Conference on Artificial Intelligence
2005 · 5.776 Zit.
Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI)
2018 · 5.478 Zit.