Tiny Eats: Eating Detection on a Microcontroller
Preprint
Abstract
There is a growing interest in low power highly efficient wearable devices for automatic dietary monitoring (ADM) [1]. The success of deep neural networks in audio event classification problems makes them ideal for this task. Deep neural networks are, however, not only computationally intensive and energy inefficient but also require a large amount of memory. To address these challenges, we propose a shallow gated recurrent unit (GRU) architecture suitable for resource-constrained applications. This paper describes the implementation of the Tiny Eats GRU, a shallow GRU neural network, on a low power micro-controller, Arm Cortex M0+, to classify eating episodes. Tiny Eats GRU is a hybrid of the traditional GRU [2] and eGRU [3] to make it small and fast enough to fit on the Arm Cortex M0+ with comparable accuracy to the traditional GRU. The Tiny Eats GRU utilizes only 4% of the Arm Cortex M0+ memory and identifies eating or non-eating episodes with 6 ms latency and accuracy of 95.15%.Keywords
All Related Versions
This publication has 13 references indexed in Scilit:
- EarBitProceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 2017
- Toward a Wearable Sensor for Eating Detection2017
- Unobtrusive and Wearable Systems for Automatic Dietary MonitoringIEEE Transactions on Biomedical Engineering, 2017
- A Novel Chewing Detection System Based on PPG, Audio, and AccelerometryIEEE Journal of Biomedical and Health Informatics, 2016
- Unobtrusive Sensing and Wearable Devices for Health InformaticsIEEE Transactions on Biomedical Engineering, 2014
- A novel approach for food intake detection using electroglottographyPhysiological Measurement, 2014
- A Survey on Human Activity Recognition using Wearable SensorsIEEE Communications Surveys & Tutorials, 2012
- Exploiting visual quasi-periodicity for real-time chewing event detection using active appearance models and support vector machinesPersonal and Ubiquitous Computing, 2011
- Self-Monitoring in Weight Loss: A Systematic Review of the LiteratureJournal of the American Dietetic Association, 2011
- Nutrition Research: An Investment in the Nation's Health2009