A Random Forests-based approach for estimating depth of human body gestures using a single video camera

An efficient Random Forests (RF)-based method for estimating the depth of human-body gestures in real-time using a single video camera is proposed. The potential application of the proposed method is in computer video games that include a 3D gesture interaction system, where human-body movements are translated to computer commands and interact directly with virtual content. The main advantage of our approach is its simplicity and cost effectiveness, since it eliminates the need for designing and implementing depth sensing cameras or 3D video cameras on computers. Performance evaluations show that our approach results in estimating very realistic depth of human-body gestures on real time basis.