Human body posture real-time estimation method based on RGB-D image feature fusion

Abstract

The invention discloses a human body posture real-time estimation method based on RGBD image feature fusion. According to the method, a method based on event triggering is adopted to obtain human bodyposture fusion estimation of depth and color images. According to the invention, through fusion of color and depth image features, error identification of a hand-held tool is effectively reduced, andthe robustness of illumination change is improved; and in order to solve the problem that human body joint point motion modeling is difficult, a BP neural network is used for carrying out human bodyjoint point motion modeling, so that the uncertainty of each joint point motion model is reduced, and the accuracy and robustness of human body posture estimation are effectively improved.

Description

Technical field[0001]The invention belongs to the field of human body posture estimation, in particular to a real-time estimation method of human body posture based on RGB-D image feature fusion.Background technique[0002]Human body pose estimation has a wide range of applications in human-computer interaction, video surveillance, assisted driving, medical motion analysis, games and sports, etc., especially in the field of human-computer interaction. Human body pose estimation is the basis for recognition of human action and behavior. It can recognize human behaviors, and complete action imitation or make response actions play a vital role.[0003]In recent years, with the rise of deep learning, the research and development of human pose estimation technology based on color images has shifted from traditional target tracking methods to target tracking methods based on deep learning. The use of convolutional neural network (CNN) method can effectively extract 2D human pose information f...

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the disadvantages of misrecognition of hand-held tools in the human body pose estimation method based on depth images and the human body pose estimation method based on color images is easily affected by illumination changes and has poor robustness, the present invention provides a human body with RGB-D image feature fusion Real-time attitude estimation method; at the same time, the use of BP neural network to model the motion of human joint points reduces the uncertainty of the motion model of each joint point and effectively improves the accuracy and robustness of human body attitude estimation

Images