Complex dynamic gesture recognition method based on Leap Motion

Abstract

The invention relates to a complex dynamic gesture recognition method based on Leap Motion, and belongs to the field of artificial intelligence and man-machine interaction. According to the method, static gesture recognition and continuous track recognition are used for complex dynamic gesture recognition, hand information in the teaching process of a user is captured through a somatosensory sensor, a support vector machine and a feature vector extraction mode based on learning are adopted for static gesture learning, and static gestures in the teaching process are all marked as instruction states. For the static gesture in the instruction state, information of the vertex of the distal bone of each finger and the central point of the palm is extracted and continuous dynamic track information is generated for learning. The complex dynamic gesture frame is decomposed by frame, and the instruction is identified after judging whether the complex dynamic gesture is the instruction gesture.According to the method, the accuracy of dynamic gesture recognition is greatly improved, the requirement for the complexity of dynamic gestures is lowered, and the man-machine interaction process ismore friendly and more natural on the basis of the visual acquisition equipment.

Description

technical field [0001] The invention belongs to the fields of artificial intelligence and human-computer interaction, and in particular relates to the realization of human-computer interaction functions based on somatosensory sensors, that is, a complex dynamic gesture recognition method based on Leap Motion. Background technique [0002] The hand is the most perfect tool formed by human beings in the long-term evolution process. As a means of communication that was first used by humans and is still widely used today, gestures have been endowed with a lot of special meanings, carrying more than 90% of body language, and are the most important way for humans to interact with robots. The ability to quickly and accurately recognize gesture commands is of great significance to the development of robot control. Gesture recognition research is a hot spot in recent years. At present, relying on the combination of sensors and artificial intelligence algorithms is the main method of...

AI Technical Summary

Problems solved by technology

[0005] Although there are many advantages in splitting dynamic gestures into static gestures for frame-by-frame recognition through visual sensors, there are still problems in the complexity and integrity of gestures because the essence is still static gesture recognition.

Different from contact sensors, complex dynamic gesture recognition based on visual sensors faces the following three problems: (1) How to effectively improve the complexity of recognizable gestures; (2) How to fully understand the continuous dynamic gestures of the teacher; ( 3) How to choose appropriate eigenvalues ​​in the process of gesture recognition

Images