计算机科学 ›› 2009, Vol. 36 ›› Issue (7): 295-297.doi: 10.11896/j.issn.1002-137X.2009.07.073

• 图形图像及体系结构 • 上一篇    下一篇

基于人体特征三维人体模型的骨架提取算法

陈国栋,李建微,潘林,余轮   

  1. (福州大学计算机图像图形研究所 福州350002)
  • 出版日期:2018-11-16 发布日期:2018-11-16
  • 基金资助:
    本文受国家自然科学基金(60672146)和卫生部科学研究基金(WKJ2005-2-004)资助。

Algorithm for Extracting Skeleton of 3D Human Body Model Based on Body Characteristic

CHEN Guo-dong,LI Jian-wei,PAN Lin,YU Lun   

  • Online:2018-11-16 Published:2018-11-16

摘要: 实现骨骼动画的一个前提是获取人体模型的骨架,现有的骨架提取算法不是计算复杂度高,就是提取准确度不高,或者需要手工干预。提出一种基于人体特点和黄金比例律的人体模型骨架提取算法,首先对模型进行精简,然后根据人体的特点与黄金比例律确定模型关节点的大概位置,在此基础上对模型进行分割。由于人体存在个体差异且姿势也可能不一致,采用测地距离方法对关节点的位置进行修正,确定其位置。与现有的算法相比,本方法效率高,同时实验显示本算法具有更好的骨架提取效果。

关键词: 算法,三维人体模型,骨架提取,黄金分割律

Abstract: Extracting skeleton of 3D human body model is very important for animation of 3D characters. The existing algorithms are complicated or interactive. We presented a algorithm that is based on the trait of human body and golden section. Our algorithm consists of four steps:1)transformation into a low polygon model from original modcl;2) estimate the position of joint on the base of trait of human body and golden section; 3) segmentation of human body; 4) amend the position using geodesic distance. The experimental results show that our method is easier to implement and faster in computation than other methods.

Key words: Algorithm, 3D human body model, Skclcton extraction, Golden section

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!