Gesture framework construction method, device and equipment, and storage medium

Abstract

The invention discloses a gesture framework construction method, a gesture framework construction device, gesture framework construction equipment and a storage medium and belongs to the field of gesture recognition. The method comprises the steps of inputting a depth picture to a gesture segmentation neural network, and outputting the depth picture to a hand confidence region, wherein the depth picture includes a depth value of each pixel point, and the depth value is used for representing a distance between an object corresponding to the pixel points and a depth camera; extracting a hand image from the depth picture according to the depth value of each pixel point in the hand confidence region; inputting the hand image to a gesture recognition neural network to obtain a three-dimensionalcoordinate of each joint of a hand; and constructing a gesture framework in a three-dimensional coordinate according to the three-dimensional coordinate. Compared with gesture interaction in a specified region, in the embodiment, the hand image is extracted by virtue of the hand confidence region output from the gesture segmentation neural network, a user can perform gesture interaction at any region in front of the depth camera, thereby being beneficial to improving the gesture recognition image and the accuracy rate for construction of a gesture framework and an interaction effect of subsequent gesture interaction.

Description

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine

Login to view more

AI Technical Summary

Benefits of technology

This patented method uses deep learning techniques for identifying areas where users interact with their hands during gestures or movements. It involves extracts important feature points from an entire scene's texture map beforehand, allowing it to be processed further without affecting its appearance. These methods are useful because they allow us to create highly accurate models of how humans behave when we touch them over different surfaces like tables top faces or walls.

Problems solved by technology

This patented describes how gestures can be used by users interact with their devices remotely without being able to recognize them accurately due to limitations such as lack of movement during interactions between different parts of the device (such as making sure that they are looking right).

Images