Detection of Face Swapped Deep Fake Videos

The Deepfake Detector is a novel application designed to enhance digital media integrity by identifying manipulated videos. This project centers on the creation and deployment of an advanced system that continuously analyzes uploaded video content to detect deepfakes. The system employs a deep learning model trained on real and fake video data, utilizing facial recognition and temporal analysis techniques. If a video is determined to be manipulated, the system informs the user with a confidence score and visual indicators, mitigating potential risks associated with deceptive media. The project also integrates with a user-friendly web interface and leverages Firebase for secure user authentication, ensuring accessibility and data protection. Furthermore, the system is engineered for efficient processing and ease of use, offering a practical solution for individuals and organizations concerned about the authenticity of video content. By improving the detection of manipulated media, this application represents a significant step toward ensuring trust in digital information.