Hand Tracking Method for Arbitrary Hand Shape

Abstract

The invention provides a tracing method of an arbitrary hand-shaped human hand, comprising the following steps of: acquiring an initialized background image; acquiring a human hand tracing initial position, and acquiring a tracing interested region according to the human hand tracing initial position; acquiring image interframe difference motion information, and acquiring an initial position region of a current human hand according to the motion information in an interested region position; updating background image information according to a preset formula; acquiring a human hand division image according to the updated background image information; and calculating the human hand position after division according to the human hand division image, and acquiring a new interested region according to the human hand position.

Description

technical field [0001] The invention relates to gesture interaction technology, in particular to a method for tracking human hands with arbitrary hand shapes. Background technique [0002] With the continuous development of computer vision technology, gesture interaction technology has been gradually popularized and applied. The core of gesture interaction based on computer vision is to use computer vision, image processing and other technologies to process the video sequences collected by image acquisition equipment, to obtain an understanding of gesture control behavior, and to respond accordingly. [0003] Hand shape tracking based on computer vision is a key technology in gesture human-computer interaction. Many scholars at home and abroad have achieved certain results in this field. The existing tracking methods can be roughly divided into: (1) based on classifiers and Fixed hand shape detection and tracking method based on particle filter; (2) Motion deformable hand s...

AI Technical Summary

Problems solved by technology

The former can get a better tracking effect and accuracy rate under a fixed hand shape, but the big defect is that the hand shape is fixed; the latter can overcome the problem of hand shape change to a certain extent, but the speed of change and movement is not very fast, and the accuracy not tall

This makes gestures have great limitations in human-computer interaction applications

Excessive restrictions on the shape and movement of human hands will bring fatigue and cumbersome problems in gesture control, which cannot meet the original intention of natural and comfortable gesture control.

Images