DeepFakeGuard- A Deepfake Detection Website Using Machine Learning

DOI

10.2991/978-94-6463-948-3_44How to use a DOI?

Keywords

EfficientNet-B0; Computer Vision; Deepfake Detection; media integrity

Abstract

The rapid advancement of artificial intelligence and synthetic media generation has amplified the threat of deepfakes, raising serious concerns about online authenticity and public trust. Deepfakes, often created using GANs and autoencoders, produce highly realistic images and videos that are increasingly difficult for humans to distinguish from authentic content. To address this challenge, we propose DeepFakeGuard, a deepfake detection framework built on the EfficientNet-B0 architecture. Unlike approaches that depend heavily on generic large-scale pretrained models, our system is fine-tuned on the target dataset to capture dataset-specific manipulation artifacts. EfficientNet’s compound scaling strategy and optimized convolutional design enable the extraction of subtle inconsistencies introduced during synthetic media generation. Furthermore, the model is integrated with a Flask-based interface, supporting real-time detection of manipulated images and videos during uploads. Experimental evaluation on the DFDC dataset demonstrates that DeepFakeGuard achieves 97% accuracy and an F1-score of 0.97, outperforming several state-of-the-art models while being computationally efficient. This balance of accuracy, speed, and usability highlights its potential for applications in education, online content moderation, and digital forensics.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Makarand Upkare
AU  - Tejas Gogawale
AU  - Vidhi Hadoltikar
AU  - Om Gurao
AU  - Veer Hajari
AU  - Shrisha Goski
PY  - 2026
DA  - 2026/01/06
TI  - DeepFakeGuard- A Deepfake Detection Website Using Machine Learning
BT  - Proceedings of the International Conference on Sustainable Innovation with Artificial Intelligence and Machine Learning 2025 (ICSIAIML 2025)
PB  - Atlantis Press
SP  - 626
EP  - 640
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-948-3_44
DO  - 10.2991/978-94-6463-948-3_44
ID  - Upkare2026
ER  -