Food Intake Actions Detection: An Improved Algorithm Toward Real-Time Analysis

Abstract

With the increase in life expectancy, one of the most important topic for scientific research, especially for the elderly, is good nutrition. In particular, with an advanced age and health issues because disorders such as Alzheimer and dementia, monitoring the subjects’ dietary habits to avoid excessive or poor nutrition is a critical role. Starting from an application aiming to monitor the food intake actions of people during a meal, already shown in a previously published paper, the present work describes some improvements that are able to make the application work in real time. The considered solution exploits the Kinect v1 device that can be installed on the ceiling, in a top-down view in an effort to preserve privacy of the subjects. The food intake actions are estimated from the analysis of depth frames. The innovations introduced in this document are related to the automatic identification of the initial and final frame for the detection of food intake actions, and to the strong revision of the procedure to identify food intake actions with respect to the original work, in order to optimize the performance of the algorithm. Evaluation of the computational effort and system performance compared to the previous version of the application has demonstrated a possible real-time applicability of the solution presented in this document.

Contactless Drink Intake Monitoring Using Depth Data

“…Chang et al showed that these two could be combined to first detect the hand holding the cup, then detect the activity such as drinking or spoon holding [7]. Due to the common problem of occlusion with vision based approaches, Cippitell et al used a top down orientation with the Kinect placed on the ceiling to capture a meal intake [8], [9], [10]. Since the Microsoft Kinect does not function properly with a topdown orientation, they developed algorithms using a Self-Organized Map algorithm to track the person's movements.…”

“…However, the dataset was limited, as it only included images of mealtime gestures. This work was expanded in [9] to improve the real-time capabilities which created a more automatic identification of the first and last frame of a food intake event. Cunha et al placed a Kinect in front of 3 elderly users consuming a meal.…”

Empowering Health in Aging: Innovation in Undernutrition Detection and Prevention Through Comprehensive Monitoring

“…The SNAQ score was weakly correlated with the MNA-SF and MNA-LF scores (r=0.392 and r=0.380, respectively, p<0.0001 for both). Gambi, Ricciuti and De Santis (2020) proposed an application for monitoring an individual's eating habits during a meal, with the eating actions being assessed through in-depth image analysis. To enhance algorithmic performance and efficiency, the author's concept involves automating the process to identify various food-related actions.…”