Unlock instant, AI-driven research and patent intelligence for your innovation.

Skeleton optimization method for continuous frames of human body based on l0 minimization

An optimization method and minimization technology, applied in image analysis, image enhancement, instruments, etc., can solve problems such as branch errors, joint fractures, and bone point overlap

Active Publication Date: 2018-11-30
BEIJING UNIV OF TECH
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In view of the above reasons, the present invention proposes a method based on L 0 Minimized skeleton optimization method for continuous frames of the human body. The main technical problems to be solved include optimizing and restoring the bone point position extraction abnormalities, branch errors and excessive unsmoothness in the continuous motion human skeleton, and solving the bone points that appear in the same frame. The problem of position coincidence and joint fracture, in order to obtain a more uniform and complete bone model

Method used

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
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Skeleton optimization method for continuous frames of human body based on l0 minimization
  • Skeleton optimization method for continuous frames of human body based on l0 minimization
  • Skeleton optimization method for continuous frames of human body based on l0 minimization

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0051] We have carried out a large number of experiments on the optimized model in the present invention, collected multi-temporal images of the human body in different movements, and reconstructed a dense human body point cloud model. After extracting the skeleton data of each frame of the human body, we use the parameters in Table 1 to optimize and get good results.

[0052] Table 1.L 0 Optimize the parameters in the model

[0053]

[0054] Figure 7 to Figure 10 It shows our multiple sets of experimental results, and it can be seen that for different human bodies and different sports, the optimization model in the present invention has achieved better results. Figure 7 Shows three frame skeletons with abnormalities before optimization. The optimization results show that the bone obtained through the optimization model in the present invention will consider the bone points of adjacent frames, so it has a strong ability to repair bone loss and abnormal position deviati...

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

PUM

No PUM Login to View More

Abstract

The invention discloses an L0 minimization-based human body continuous frame skeleton optimization method, belongs to the field of graphical representation methods, and aims at optimizing and restoring conditions such as skeleton point position extraction abnormities, branch errors and excessive un-smoothness in continuous motion human body skeletons and solving the problems of skeleton point position superposition and joint fracture in a same frame so as to obtain more uniform and integrated skeleton models. The method comprises the following steps of: firstly acquiring images of a human body in multi-view movement, reconstructing a dense point cloud model by using each view angle of each frame, and extracting a skeleton model by using an L1 mid-value skeleton method for the point cloud model of each frame; blocking the human body skeleton of each frame, sampling skeleton point of each block, constructing a continuous frame human body skeleton sequence, and establishing a sparse optimization model considering relevance of adjacent frames; and finally adding regular terms of adjacent skeleton points of a same frame in the optimization model to ensure that the optimized skeleton points are more uniform.

Description

technical field [0001] The present invention is applicable to a graphic representation method, in particular to an optimization method for extracting bones from a human body in consecutive frames. Background technique [0002] The skeletal framework extracted from the 3D model can be considered as a very intuitive and efficient representation of the model, which has strong practical significance for analyzing and manipulating the 3D model. Especially for human models, skeletal frameworks can be used for pose estimation, human modeling and manipulation. [0003] At present, there are already some techniques that can directly extract skeletons from incomplete and noisy point cloud models, such as skeleton model extraction based on the concept of generalized rotational symmetry axes, L 1 Central bone extraction technology. But for continuously moving human models, these methods do not take into account the interrelationships between the motion models of each frame. If only t...

Claims

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

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/246
CPCG06T2207/10028
Inventor 尹宝才沈伯伟张勇王少帆孔德慧
Owner BEIJING UNIV OF TECH