Inspiration
We were inspired to build a solution that helps people verify the authenticity of videos and prevent the spread of harmful content.
What it does
Our platform allows users to upload a video and instantly check whether it is real or AI-generated (deepfake)
How we built it
Frontend: HTML, CSS, JavaScript (or React) Backend: Python with frameworks Django AI Model: Deep learning models (CNN/LSTM) trained on deepfake datasets Libraries: OpenCV, TensorFlow
Challenges we ran into
Finding high-quality datasets for training Handling large video files efficiently Improving accuracy (reducing false positives/negatives) Optimizing performance for real-time detection
Accomplishments that we're proud of
Built a working AI-based detection system Achieved good accuracy on test samples Created a simple and intuitive user interface Addressed a real-world problem with social impact
What we learned
Practical implementation of Artificial Intelligence and deep learning Working with video processing and computer vision Importance of ethical AI and responsible tech
What's next for DeepFake AI video detector
Improve model accuracy with larger datasets Add real-time detection (live video/webcam) Develop a browser extension for instant checks Provide API integration for companies and platforms Add audio deepfake detection
Log in or sign up for Devpost to join the conversation.