https://github.com/heitan104/truesight
Detect AI-generated short-form videos using deep learning. Built with Flask, TensorFlow, and React Native.
https://github.com/heitan104/truesight
ai-detection deepfake expo flask hackathon keras mobile-app opencv react-native spurhacks tensorflow video-analysis
Last synced: 3 months ago
JSON representation
Detect AI-generated short-form videos using deep learning. Built with Flask, TensorFlow, and React Native.
- Host: GitHub
- URL: https://github.com/heitan104/truesight
- Owner: Heitan104
- Created: 2025-06-21T04:02:38.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2025-07-02T03:35:23.000Z (12 months ago)
- Last Synced: 2025-07-02T04:32:31.340Z (12 months ago)
- Topics: ai-detection, deepfake, expo, flask, hackathon, keras, mobile-app, opencv, react-native, spurhacks, tensorflow, video-analysis
- Language: TypeScript
- Homepage:
- Size: 66.4 MB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# TrueSight
TrueSight is an AI-powered content authenticity detector created for Spurhacks2025. It analyzes shared TikTok/Reel links in real-time and instantly notifies users whether the content is AI-generated—without interrupting their scrolling session. With TrueSight, you can stay informed about the content you consume while maintaining your social media flow.
### Inspiration
We built TrueSight to combat the rising wave of AI-generated misinformation on short-form video platforms. Our goal is to empower users to identify synthetic content effortlessly, promoting digital literacy without disrupting their browsing experience.
### Features
- **Instant AI Detection**: Share a TikTok/Reel link and receive a notification with an AI probability score (e.g., "92% likely AI-generated").
- **Explainable Results**: Clear reasons for the detection (e.g., "Unnatural facial movements, synthetic voice patterns").
- **Seamless Integration**: Works entirely in-app—no switching screens or losing your place.
- **History Log**: Optional saved results for reviewed content (future update).
### How It Works
1. **Share a Link**: User shares a TikTok/Reel URL to TrueSight (via app/extension).
2. **AI Analysis**: Our model processes the video’s visual/audio features.
3. **Smart Notification**: Returns a percentage score and detection rationale directly to the user’s device.
### How to Use
1. **Mobile App**:
- Open TrueSight and enable notifications.
- Share TikTok/Reel links directly to the app.
### Setup and Installation
To run TrueSight locally:
```bash
git clone https://github.com/your-username/TrueSight.git
cd TrueSight
pip install -r requirements.txt # AI dependencies
npm install # Frontend dependencies