Demo: Smartwatch based food diary and eating analytics

• Main Authors: SEN, Sougata, SUBBARAJU, Vigneshwaran, MISRA, Archan, LEE, Youngki, BALAN, Rajesh Krishna
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2016
• Subjects: Computer Sciences; Numerical Analysis and Scientific Computing; Software Engineering
• Online Access: https://ink.library.smu.edu.sg/sis_research/3281; https://ink.library.smu.edu.sg/context/sis_research/article/4283/viewcontent/9924678.pdf
• Institution: Singapore Management University

Monitoring an individual’s daily dietary intake can provide various insights regarding the health of the individual. Applications such as My Fitness Pal exists, which allows individuals to monitor all items that the individual consumed. However, manual monitoring can be labour intensive. To overcome this limitation, wrist worn sensor based eating habit monitoring has been studied by various researchers. These systems can detect eating gesture, but they cannot determine what is being eaten. We have built a system which can (i) detect eating gesture using the smartwatch's inertial sensors (ii) use the smartwatch's camera to capture images of food consumed at opportune moment and (iii) present the images to the user at the end of day. Key challenges in building the system are : (i) limited battery (ii) diversity in eating style and (iii) latency in camera triggering.