Human body posture recognition method and system

Abstract

The invention discloses a human body posture recognition method. The recognition method comprises the following steps: acquiring a plurality of human body images of a current frame under different visual angles; establishing a three-dimensional human body model by adopting a convolutional neural network algorithm according to the plurality of human body images; clustering adjacent pixels on each human body image, and obtaining a plurality of pixel blocks; establishing a two-dimensional pixel block model of each pixel block; optimizing the three-dimensional human body model according to the two-dimensional pixel block model of each pixel block, and obtaining an optimized three-dimensional human body model; and determining the human body posture of the current frame according to the optimized three-dimensional human body model. According to the method, the three-dimensional human body model is established simply and quickly by using the convolutional neural network algorithm, then the three-dimensional human body model is optimized by using the human body posture in the image, and the posture recognition is performed by using the optimized three-dimensional human body model, so that the human body posture estimation accuracy is ensured, and the human body posture estimation speed is improved.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a method and system for human gesture recognition. Background technique [0002] 3D human pose estimation refers to estimating the pose of a human target from an image, video, or point cloud, and is a basic task in 3D research around the human body. 3D human pose estimation is an important prerequisite for 3D human reconstruction, and it can also be an important source of motion in human motion driving. At present, there are mainly two implementation methods for obtaining human body pose. 1. Use the neural network to achieve the goal of human body pose estimation in this scene by training on a specific data set. This method requires a large amount of manually labeled data to train the neural network, and at the same time, the accuracy of this method is low. 2. Establish a human body model and make the model fit the human body on the picture. This method relie...

AI Technical Summary

Problems solved by technology

This method relies on the establishment of the human body model, the process of establishing the human body model is relatively complicated, and the speed of this method is relatively slow.

Images