Passive Observer of Activities for Aging in Place Using a Network of RGB-D Sensors

Figure 12

Similar to Figure 8, this figure shows two views of the calibrated virtual scene of the kitchen based on the merged point clouds of the two depth sensors. Both views show the segmented portion of the point clouds associated with the detected objects after the older adult has left the monitoring area. Virtual reconstruction size of the point cloud is based on the physical sizes of the objects. The detected objects in the RGB image are used to segment the associated point cloud for determining their approximate position with respect to the real and virtual scenes. For example, the orange points represent point clouds associated with the final position of the oven. Although the movement trajectory information is not used in this paper, the figure shows also a sample movement trajectory of older adult in the monitoring area.

(a)

(b)