A vision-based framework for building telecare and telerehabilitation services

Experimental work: LOWER TRAIN WOMEN. The next video was processed by means of a neural network with a data set used for learning and training to lower train, and the configuration of the neural network is based on 18 inputs, 1 output and 10 hidden layers.

 

Experimental work: LOWER TRAIN MEN. The next video was processed using the same configuration used for previous video.

 

Experimental work: UPPER TRAIN WOMEN. The next video was processed by means of a neural network with a data set used for learning and training to upper train, and the configuration of the neural network is based on 15 inputs, 3 outputs and 12 hidden layers.

 

Experimental work: UPPER TRAIN MEN. The next video was processed using the same configuration used for previous video.

 

Experimental work: FALL DETECTION AND ALARM. The next video was recorded with a camera located in front of the subject, i.e. both vertical and horizontal angles with respect to the subject were 0º.