InfraVision AI is a mobile-first AI-powered tool that revolutionizes infrastructure inspection. It leverages cutting-edge computer vision and AI to detect structural damage, generate 3D visualizations, and provide safety reports in real time.
-Real-Time Damage Detection: Use your mobile camera to scan structures and detect cracks, corrosion, or deformation.
-AI-Powered Risk Analysis: LLM Rage database with Pinecone Vector DB using Gemini 1.5 Flash analyzes and predicts infrastructure health status.
-Safety Report Generator: Automatically generates detailed safety reports for records or government use.
-Augmented Reality Mode (Coming Soon): AR view highlighting stress zones in infrastructure.
-Voice Input & Multilingual Support: Speak to the app and get results in your preferred language.
-User Authentication: Powered by Clerk for secure and seamless login.
To enable smart, proactive infrastructure inspection using AI for safer cities, reduced manual labor, and cost-effective maintenance.
To become the go-to AI platform for infrastructure health diagnostics, serving governments, engineers, and developers globally.
-
Real-time AI scanning with LLM Rage DB and Pinecone
-
Mobile-first inspection from anywhere
-
User-friendly UI & AR-ready interface
-
Generates official-grade safety reports
- High computational demand for real-time 3D generation
- Depends on external APIs (Gemini, Clerk)
- Limited to camera-based visible surface detection
- Use in disaster zones, post-earthquake scans
- Government tie-ups for urban safety
- Expand into bridge, railway, and dam inspections
- Data analytics for predictive maintenance
- Strong competitors in AI infrastructure space
- Data privacy concerns while capturing site visuals
- AI hallucinations in poor lighting or camera angles
- Frontend: React.js, Material UI, CSS Modules
- Authentication: Clerk.dev
- AI/ML: Google Gemini 1.5 Flash Pinecone VectorDB Rage Database
- Hosting: Netlify / Vercel (Frontend), Firebase (optional)
/infravision-ai
│── /public # Static assets like images, manifest, etc.
│── /src
│ ├── /app # Main app logic and views
│ ├── /components # Reusable UI Components (Navbars, Cards, etc.)
│ ├── /lib # Utility functions or constants
│ ├── /pages # Page-level components
│ ├── /styles # Global or component-specific styling
│ ├── /utils # Helper functions and modules
│ ├── /auth # Clerk authentication setup
│ ├── pinecone.js # Pinecone vector search setup
│ ├── langchain.js # Langchain configuration
│ └── index.js # App entry point
│
├── .env # Environment variables
├── package.json # Project dependencies and scripts
└── README.md # Project documentation
- Node.js 18+
- React Project + Rage Database
- (Optional) Google Gemini API Key for live scans
git clone https://github.com/debmallya/infra-vision-ai.git
cd infra-vision-ai
npm install🧬 Environment Variables Create a .env.local file:
REACT_APP_CLERK_PUBLISHABLE_KEY=<your-clerk-publishable-key>
REACT_APP_CLERK_SECRET_KEY=<your-clerk-secret-key>
REACT_APP_GEMINI_API_KEY=<your-gemini-api-key> # optional
npm run dev
Click "Explore Now" on the landing page.
Sign up or log in with Clerk authentication.
Navigate to the camera scan dashboard.
Point camera towards structure – AI detects damage.
View visualized 3D model and safety report.
Export report or continue scanning.
We welcome all contributions! Fork the repo, make your changes, and submit a pull request. Let’s build safer cities together.
⭐ Star this repo if you like the project!