Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{180016,
        author = {Rupesh Kumar Sah and Prakash Prajapati and Prateek Rajput and Priyanshu Proji and Palak Shandil},
        title = {Deepfake detection using CNN},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {1},
        pages = {866-874},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=180016},
        abstract = {The Deepfake Detection Project takes on the important challenge of deepfake video technology, which is a domain of artificial intelligence that generates incredibly realistic, yet fabricated, videos. It utilizes a cutting-edge approach by combining Convolutional Neural Networks (CNNs) to extract spatial features, and Gated Recurrent Units (GRUs) to analyze temporal features. By combining these two powerful deep learning methods, it allows for the detection of small inconsistencies in facial expressions, lighting and movement patterns across video frames that would otherwise not be caught to robustly and accurately identify manipulated media. The system is deployed via a Flask-based backend, and TensorFlow for the model deployment. The frontend, minimalistic and user-friendly, allows interaction with the system. Furthermore, the system can easily run on local computers or cloud infrastructure, affording flexibility and scalability for the use in journalism, security, and digital content verification. Building upon previous systems and addressing their limitations, along with incorporating temporal modelling, the impact of this project provides important advances in the identification of sophisticated deepfakes, hone in the accountability and integrity of digital media.},
        keywords = {},
        month = {June},
        }