Our advanced machine learning model analyzes facial images and decides whether they are real or have been altered
In an era of synthetic media, discerning truth from fiction is increasingly challenging
Deepfakes can spread false narratives, manipulate public opinion, and undermine trust in institutions.
Synthetic identities and voice cloning enable sophisticated financial and social engineering scams.
We empower users with tools to critically evaluate digital content and combat digital deception.
A streamlined, multi-stage AI pipeline built for interpretability, speed, and real-world reliability.
The system first locates all visible faces using the Haar Cascade detection algorithm. Each detected face is cropped so the analysis focuses only on regions where manipulation typically occurs.
Final predictions combine results from multiple specialized models with meta-learning to achieve [RESULT] on our test datasets. Results include confidence scores and visual explanations.
Electrical & Computer Engineering Seniors & Faculty Advisor
Team Lead / Machine Learning Engineer
Led the team and developed the CNN model architecture, training pipeline, and system integration.
Data Research Analyst
Researched data sources, and ensured data quality.
Data Engineer
Managed datasets and implemented the feature extraction process to prepare inputs for model training.
Web Developer
Designed and built the website for showcasing and deploying the project model.
Dr. Picone
All uploaded media is immediately deleted.
Our model achieves [RESULT] ]on test datasets, but we clearly communicate limitations and potential biases.
This tool is for demonstration purposes only. We prohibit harassment, discrimination, or misuse.
Explore our technical implementation and research