摘要：In the present fast-moving and tech-savvy world, virtual environments and virtual elements which can fraternize with real-world objects have become common. Robots or machines that are palpable can be handled and supervised through mere gestures using advanced technology. Most gesture recognition algorithms focus on bodily gestures originating from the face or hands (although research is being done in gait recognition as well). Not only are the face and hands easier body parts to detect (imagine trying to create an object detector to detect the presence of a torso, for example), but also they can easily convey a huge amount of information with only simple gestures. The human face can convey various emotions such as happiness, pain, anguish, and elation, while the hands can quickly convey more complex concepts using sign language. The overall goal of hand gesture recognition is to bridge the gap between our human body language and computer interpretation, allowing us to interact with our computers more richly. Perhaps in the future, we will no longer need keyboards and mice as input! However, we are far from this level of gesture recognition, and much more research needs to be performed. As the name suggests, hand gesture recognition involves the movement of hands and fingers. The computer then applies mathematical models to interpret these gestures and interact with the human. In this project motion detection and background subtractor are applied as a precursor to gesture recognition. 'Malima et al.' method is used to build a computer vision system that adds two numbers together based on hand gestures.
关键词：OpenCV;Gesture Recognition;Hand Gesture Recognition;Appearance Based Model;Simple Camera;Segmentation
