Personalized Dietary Self-Management using Mobile Vision-based Assistance

Publication from Digital

Waltner, Georg and Schwarz, Michael and Ladstätter, Stefan, Weber, Anna and Luley, Patrick, Lindschinger, Meinrad and Schmid, Irene and Scheitz, Walter and Bischof, Horst and Paletta, Lucas

Proc. ICIAP 2017 Workshops, Springer LNCS, 3rd International Workshop on Multimedia Assisted Dietary Management, MADIMA 2017, in conjunction with the 19th International Conference on Image Analysis and Processing, September 12, 2017, Catania, Italy. , 1/2017


Daily appropriate decision making on nutrition requires application of knowledge where it matters, and being adjusted to the individual requirements.We present a highly personalized mobile application that assists the user in appropriate food choices during grocery shopping, while simultaneously incorporating a personalized dietary recommender system. The application can be used in video based augmented reality mode, where a computer vision algorithm recognizes presented food items and thus replaces tedious search within the food database.The recognition system employs a shallow Convolutional Neural Network (CNN) based classier running at 10fps. An innovative user study demonstrates the high usability and user experience of the application. The vision classier is evaluated on a newly introduced reference image database containing 81 grocery foods (vegetables, fruits).