Generating recipes based on ingredients
Generate recipes based on upload images

Inspiration

We wanted to improve our cooking skills and reduce food waste by using all available ingredients effectively. As students who love cooking and savoring good food but don't have much time to spend in the kitchen every day, we needed a way to whip up delightful meals efficiently without compromising on taste and satisfaction.

What it does

We developed RecipeMakerAI using a combination of React for the frontend, Node.js for the backend, and integrated various APIs to enhance functionality. The Edamam API powers our ingredient-to-recipe search, while TensorFlow's pre-trained models enable image recognition for ingredient detection. We also used React Router for navigation and Axios for API requests.

Challenges we ran into

We encountered several challenges, particularly with implementing the Edamam API. Initially, we attempted to use Node.js, but the API responses were only displayed on the console. We switched to pure JavaScript for better integration. Additionally, we struggled with finding an optimal image classification model and ultimately chose TensorFlow's pre-built model after extensive research.

Accomplishments that we're proud of

We are proud to have successfully integrated image detection and API implementation. Despite the hurdles, we managed to get the image detection working and effectively utilized the APIs to enhance the app's functionality.

What we learned

Throughout this project, we learned a great deal about API implementation and identifying suitable databases for our needs. We also explored using OpenAI to display recipe preparation times. Furthermore, we deepened our understanding of TensorFlow's pre-built models and their capabilities. Our skills in React were strengthened as we built multiple components, used React Router DOM, and integrated various APIs to create a comprehensive app with both frontend and backend features.

What's next for RecipieMaker

Looking ahead, we plan to implement more features to assist users with disabilities and dietary restrictions. We aim to improve our image classification system by using a larger database for more accurate food identification. Additionally, we want to expand our model to support real-time object detection via video and provide more options to filter recipes based on cuisine, meal type, allergy, or dietary restrictions.

We also envision enabling users to shop for missing ingredients and find the best deals nearby. Future updates will include a trending recipes section and user reviews for generated recipes to better cater to our users' needs.

Next steps for RecipeMaker involve refining accuracy in ingredient detection. This will be achieved through fine-tuning and training our own dataset to ensure more precise and reliable results.