Dynamic Hand Gesture Recognition using cyclical patterns of hand movement and its applications ☆

This paper tackles a new prototype of dynamic hand gestures and its advantages to apply to controlling smart home appliances. The proposed gestures convey cyclical patterns of hand shapes as well as hand movements. Thanks to the periodicity of defined gestures, on one hand, common technical issues that appear when deploying the application (e.g., spotting gestures from a video stream) are addressed. On the other hand, they are supportive features for deploying a robust recognition scheme. To this end, we propose a novel hand representation in a temporal-spatial space. Particularly, the phase continuity of the gesture's trajectory is taken into account underlying the temporal-spatial space. This scheme obtains very promising results with the best accuracy rate is of 96%. The proposed techniques are deployed to control home-appliances such as lamps, fans. These systems have been evaluated in both lab-based environment and real exhibitions. In the future, the proposed method will be evaluated in term of naturalness of end-users and/or robustness of the systems.