Multi-Sensor Dietary Tracking Device with AI Food Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional self-serve food services lack the capability to provide users with instant dietary feedback, such as portion-based nutrition information and allergen alerts, especially for unpacked foods.
Innovation Solution
An AI-assisted multi-sensor device that uses machine learning models to identify food and estimate portion sizes based on images and depth data, providing users with real-time dietary feedback on a display.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If traditional self-serve food services are used, then device complexity is low, but dietary information feedback is lacking
Solution Approach 1:
The system captures images of food items, processes them through machine learning models to identify food type and estimate portion size, then provides real-time feedback to users about nutritional information, calories, and dietary recommendations. This closed-loop feedback mechanism transforms the traditional open-loop self-serve system into an informed decision-making platform.
Solution Approach 2:
The patent replaces manual food tracking and estimation with automated optical sensing and machine learning algorithms. Instead of users manually recording food intake, the system uses image recognition and depth sensing to automatically identify food items and calculate portion sizes, substituting mechanical/manual processes with intelligent automated systems.
2Measurement precision
If machine learning models are used for food identification, then measurement precision of food type is improved, but device complexity increases
Solution Approach 1:
The system divides the complex food recognition task into multiple specialized machine learning models: one model for food type identification and another for portion size estimation. This segmentation allows each model to focus on a specific aspect of food analysis, improving overall accuracy while managing computational complexity through modular architecture.
Solution Approach 2:
The system transitions from 2D image analysis to 3D depth-aware food recognition by incorporating depth images and generating 3D representations of food items. This dimensional enhancement provides more geometric and spatial information for accurate portion size estimation, improving measurement precision by adding a third dimension to the analysis.
3Measurement precision
If depth images and 3D representations are used, then portion size estimation accuracy is improved, but use of energy increases
Solution Approach 1:
The system processes depth images and 3D representations only when necessary for accurate portion estimation, rather than continuously analyzing all food items with full 3D processing. By applying complex computational methods selectively based on food type and context, the system achieves high accuracy for difficult cases while conserving energy for simpler scenarios.
Data Source
AI summary
Methods, systems, and apparatuses, including computer programs encoded on computer storage media, for dietary tracking with instant dietary feedback using a portable multi-sensor device are described. An example method may include: obtaining a first machine learning model trained based on a plurality of food images and corresponding labels; receiving a series of images from one or more cameras capturing a user taking food out of a food-serving storage; determining, using the first machine learning model based on the series of images, an identification of the food being taken by the user; determining, using a second machine learning model based on the series of images and the identification of the food, an estimated weight of the food being taken by the user; determining and displaying portion-based food information of the food being taken by the user.