Optical unmarked three-dimensional human body motion capture method

Abstract

The invention provides an optical unmarked three-dimensional human body motion capture method, solves the problem of constraint of traditional non-optical and optical marked human body motion capturemethods on human body motion, and reduces the duration of single human body motion capture to a certain extent. The method comprises the following steps: determining a human body joint number and a connection relationship among all joint points by utilizing a corresponding relationship between positions of different joint points and a real human body; performing two-dimensional joint point extraction of the multi-angle human body image and limb skeleton connection lines between joint points by utilizing a deep convolutional neural network and combining the serial number and the connection relationship to obtain two-dimensional coordinate information of the human body joint points; and drawing a human body skeleton model in a three-dimensional space by utilizing a conversion relationship between different coordinate systems and the two-dimensional coordinate information of the human body articulation points.The skeleton model reflects real human body posture motion information in the three-dimensional space, and the motion capture method is conveniently applied to the field of human body motion analysis subsequently.

Description

technical field [0001] The invention provides an optical markless three-dimensional human motion capture method, which belongs to the field of 3D human motion capture. Background technique [0002] 3D human body motion capture refers to the use of certain methods or means to obtain the three-dimensional space coordinates and three-dimensional space distribution of human joint points through sensor perception and background program calculations, and according to the interconnection relationship between each joint point, point and The connection between the points, and finally draw the human skeleton model in the three-dimensional space coordinate system, so as to achieve the purpose of human motion motion capture. [0003] At present, this research direction is a research hotspot in the field of computer vision. Different from the traditional method of human body posture detection through wearable sensors (Hu Xiaohua, Li Xiangpan. Design of wearable human body posture detect...

AI Technical Summary

Problems solved by technology

[0009] The invention provides an optical three-dimensional human body motion capture method without markers, which solves the problem of the traditional non-optical and optical marked human motion capture methods on human body movement, and reduces the number of single human motion captures to a certain extent. duration of

Images