AI-powered carbon footprint decoder that transforms receipts, bills, and daily activities into actionable environmental impact reports.
People lack visibility into the carbon impact of everyday purchases and activities. Manual carbon accounting is tedious and generic tips don't reflect actual behavior.
Upload a receipt image or describe your day in plain text. EcoMate-AI extracts activities via OCR/NLP, maps them to verified global emission factors, and generates personalized sustainability recommendations with interactive visualizations.
| Capability | Detail |
|---|---|
| Input modes | Receipt images + free-form text |
| Emission mapping | Verified global CO₂e factors (data/emission_factor.csv) |
| Output | Category breakdowns, global comparisons, ranked green tips |
| Stack | Streamlit + FastAPI + GPT-4o Vision |
- Multimodal input — upload receipt images or enter free-form text
- OCR extraction — reads items, quantities, and services from scanned receipts
- CO₂ estimation — maps activities to verified global emission factors
- Personalized tips — AI-generated suggestions ranked by impact
- Global comparison — contextualizes your footprint against regional averages
- Interactive visualizations — category breakdowns and trend charts
| Layer | Tools |
|---|---|
| Frontend | Streamlit |
| Backend | FastAPI |
| AI / OCR | OpenAI GPT-4o, Vision API |
| Data | pandas, NumPy |
| Visualization | Plotly, Matplotlib |
git clone https://github.com/ShamikOfficial/EcoMate-AI.git
cd EcoMate-AI
python -m venv venv && source venv/bin/activate # Windows: venv\Scripts\activate
pip install -r requirements.txt
cp .env.example .env # add your OpenAI API key
streamlit run app/main.pyEcoMate-AI/
├── app/
│ ├── main.py # Streamlit frontend
│ ├── api.py # FastAPI backend
│ ├── genai_model.py # GenAI inference layer
│ └── services/ # Carbon calculation logic
├── data/ # Emission factor datasets
├── docs/architecture.md # System design
└── requirements.txt
See docs/architecture.md for the full pipeline diagram.
MIT — see LICENSE.