Poor diet is the leading cause of death and disability in the world. Hence, we built AugmenTable to help those with dietary restrictions such as allergies, religious/belief diets, health diets, etc. Through Vision AI and Barcode Reading, AugmenTable cross-references ingredients lists and nutrition facts with global databases, ensuring that it delivers the most accurate results and recommendations.
AugmenTable uses AR overlays with Scandit to provide easy identification of barcodes as well as fast and obvious recommendations on whether to avoid, be cautious, or enjoy. We believe that knowing what foods are safe to eat shouldn't be a hassle or a potential hazard, so we made it as easy as possible to eat without a single worry.
Using a global database covering a wide range of food barcodes and metadata, AugmenTable delivers highly-accurate results at a very fast pace due to the lack of OCR. In addition to these results, we use generative transformer AI to provide coherent Enlgish reasoning to any recommendation to make sure you know why any recommendation was made.
- The prototype for AugmenTable was extensively tested on Android - iOS compatibility is unkown.
- This is our first
react-nativeapp, and backend logic is as always in./backend/
- Custom user profiling
- Amazon integration
- Diet plans
- Image analysis of meals without barcode with state of the art image analysis models (code already in place)
- 100% deep learning based analysis of foods and ingredients.