What can computer vision tell you about your weight?

Velardo, Carmelo; Dugelay, Jean-Luc
EUSIPCO 2012, 20th European Signal Processing Conference 2011, August 27-31, 2012, Bucharest, Romania

Extreme overweight and obesity are spreading widely like an epidemic. Populations feels the urge for preventive tools and methods to increase people self-awareness and 3D body analysis could be employed to create selfdiagnostic methods. We propose a system based on the Microsoft Kinect RGBD sensor to help people to detect weight problems and, possibly, to guide them through improvements in their lifestyle. Our application extrapolates

anthropometric measures from the body silhouette and 3D information. Measures are used to estimate body weight thanks to a statistical model trained on the data of a large medical database. Thanks to the knowledge acquired while analyzing the user, the system provides

through its interface, healthiness measures and information that are necessary for a correct lifestyle that would mitigate the effects of weight problems.


Type:
Conference
City:
Bucharest
Date:
2012-08-27
Department:
Digital Security
Eurecom Ref:
3756
Copyright:
© EURASIP. Personal use of this material is permitted. The definitive version of this paper was published in EUSIPCO 2012, 20th European Signal Processing Conference 2011, August 27-31, 2012, Bucharest, Romania and is available at :

PERMALINK : https://www.eurecom.fr/publication/3756