The accurate assessment of nutrition information is a challenging task, but crucial for people with certain diseases, such as diabetes. An important part of the assessment of nutrition information is portion estimation, i.e. volume estimation. Given the volume and the food type, the nutrition information can be computed on the basis of the food-type-specific nutrition density. Recently mobile devices with depth sensors have been made available for the public (Googles project tango platform). In this project, an app for mobile devices with a depth sensor is presented which assists users in portion estimation. Furthermore, we present the design of a user study for the app as well as preliminary results.
EatAR Tango – Portion Estimation using AR
Projektwebsite: www.oekogotschi.at Laufzeit: Jänner 2017 Team: Radomir Dinic, Michael Domhardt, Simon Ginzinger, Thomas Stütz
2017 IEEE International Symposium on Mixed and Augmented Reality (ISMAR-Adjunct), pp. 284-287, 2017.
EatAR Tango: Results on the Accuracy of Portion Estimation (2017)
Proceedings of the 19th International Conference on Human-Computer Interaction with Mobile Devices and Services, pp. 46:1–46:7, ACM, Vienna, Austria, 2017, ISBN: 978-1-4503-5075-4.
EatAR Tango: Portion Estimation on Mobile Devices with a Depth Sensor (2017)