SnapStem is a plant identification and care information app that leverages Pinecone DB for efficient vector storage and search, alongside OpenAI for identifying plants from user-uploaded photos. With the added PlantPal chatbot feature, users can get detailed care information and engage in personalized conversations about their plants.
SnapStem is designed to help plant owners identify and care for their plants by using AI. Users can upload photos of their plants, and the app will use OpenAI to identify the plant and provide care requirements. Additionally, the app includes a chatbot functionality, allowing users to ask specific questions about their plants and receive guidance.
- Camera Functionality: Upload images of plants for identification.
- Chatbot: Ask questions about plant care and health.
- Image Similarity Rating: Get a similarity score to the identified plant.
To get started with SnapStem, follow these steps:
-
Clone the repository:
git clone https://github.com/daniellem62/SnapStem
-
Navigate into the project directory:
cd SnapStem- Install dependencies:
npm install- After installation, start the development server with:
npm run devYour app will be available at http://localhost:3000.
Some stretch goals for the project are to:
- Add more plants to the database. Currently, there are a limited of number of plants in the database. Therefore, if a user tries to identify a plant that is not in the database it will not give an accurate result.
- Add a plant inventory/wishlist feature for users to track their plants.
- Implement plant care schedule tracking, with reminders for watering, fertilizing, and pruning.
- Improve plant identification accuracy by integrating a web crawler or API database for expanded care information.
- Add functionality for plant health queries (e.g., “Why is my plant's leaves curling?”).
- Implement plant disease identification via image uploads. Add a login page with user authentication.
Feel free to reach out to us with any questions, suggestions, or feedback.
Dani Holly Nasrin Louis Sam Owen
Project Link: Click here to view the project