International Journal For Multidisciplinary Research

E-ISSN: 2582-2160   •   Impact Factor: 9.24

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Call for Paper Volume 8, Issue 6 (November-December 2026) Submit your research before last 3 days of December to publish your research paper in the issue of November-December.

AI Based Deepfake Detection Using Convolutional Neural Networks and Transfer Learning

Author(s) Ms. Sakshi Dattatray Jagtap, Mr. Kaustubh Ramesh Jagdale, Mr. Nishant Shivaji Jadhav, Ms. Sayali Jaganath Patil, Dr. Nilesh Thorat
Country India
Abstract DeepShield AI is a web-based deepfake detection system designed to identify potentially manipulated facial images and videos using deep learning techniques. The system uses a Next.js frontend and FastAPI backend integrated with a PyTorch-based CNN classification pipeline. It supports image uploads, short video analysis through frame sampling, and selected live-camera frame analysis. OpenCV Haar Cascade is used for face detection and cropping. The system classifies media as REAL or FAKE based on model outputs and displays confidence scores with appropriate warnings. The project focuses on transparent detection and avoids generating artificial results when trained model weights are unavailable. The current implementation has limitations, and further model training and evaluation are required to improve reliability.
Keywords Deepfake Detection, Artificial Intelligence, Convolutional Neural Network, Transfer Learning, Deep Learning, EfficientNet-B0, Face Detection, Image Processing, PyTorch
Field Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In Volume 8, Issue 6, November-December 2026
Published On 2026-11-20

Share this