Method and system for gesture recognition

Abstract

The invention relates to a method and a system for gesture recognition. The method comprises steps of transmitting a structure light plane through an infrared structure light transmitting unit; receiving an induced image formed by infrared light signals which are reflected by a user through an infrared structure light sensor; acquiring the depth information of the induced image through a depth image processing unit; selecting and recording the information, i.e., the hand motion track information, which is close to the infrared structure light transmitting unit of the depth information through a hand target following unit; and recognizing the hand motion track information through a gesture recognition unit and determining corresponding user gesture control commands. By the aid of the method and the system, the user dynamic hand gesture can be separated from a complex background, based on space direction dividing recognition, the gesture recognition is completed, the gesture modeling and learning process are not conducted in advance and the user experiment is improved.

Description

technical field [0001] The present invention relates to human-computer interaction technology, in particular to a gesture recognition method and system. Background technique [0002] Human-computer interaction technology is a very popular research field in recent years. Mouse and keyboard are traditional human-computer interaction devices. In recent years, various new human-computer interaction methods such as touch control, voice control, and gesture control have emerged. Improve the naturalness and friendliness of the user experience. In particular, the non-contact human-computer interaction method represented by gesture control, through various sensor devices, completes the recognition process of hand movements in real time or in a short period of time, and converts them into commands that can be recognized by host devices such as computers. It is a very popular way of human-computer interaction. [0003] Gesture action recognition methods can be divided into two types:...

AI Technical Summary

Problems solved by technology

Especially for dynamic gestures, different users will have speed differences, trajectory differences, proficiency differences, etc. when performing gestures, which will lead to nonlinear fluctuations in the gesture modeling trajectory on the time axis, and this non-linear The elimination of linear fluctuations is very difficult and complicated, so the traditional dynamic recognition rate based on two-dimensional images is generally not high enough

Images