Hand gesture recognition with focus on leap motion: An overview, real world challenges and future directions

In the recent years, a steady growth of Hand Gesture Recognition (HGR) based applications has been observed. Thus, significant progress has been made in the field of hand detection and tracking. However, hand poses remain particularly challenging and can deeply affect the results due to several factors especially congenital hand problems. Besides, HGR applications performance is very dependent on the sensor used for gesture acquisition. Owing to its high performance, the Leap Motion Controller (LMC), has recently become a by default sensor for HGR systems and provide high performance in terms of recognition accuracy. This work proposes a comprehensive literature review of HGR based on different modalities with a focus on the LMC based systems. This survey retrieves the main challenges faced in literature with an investigation of the state of the art solutions to address them. Moreover, we discuss numerous existing methods for static and dynamic HGR from different modalities. We reveal remaining challenges in the literature and highlight promising future research directions.