PROJECT TITLE :

Fusion of depth ,skeleton ,and inertial data for human action recognition - 2016

ABSTRACT:

This paper presents somebody's action recognition approach by the simultaneous deployment of a second generation Kinect depth sensor and a wearable inertial sensor. Three knowledge modalities consisting of depth images, skeleton joint positions, and inertial signals are fused by utilizing 3 collaborative illustration classifiers. A database consisting of ten actions performed by six subjects is put along to hold out 2 types of testing of the developed fusion approach: subject-generic and subject-specific. The general recognition rates obtained from each varieties of testing indicate recognition improvements when fusing all the info modalities compared to the situations when knowledge modalities are used individually.