|本期目录/Table of Contents|

[1]汪亚明,张静,郑俊褒.基于加速度平滑约束的非刚体三维重建研究[J].浙江理工大学学报,2017,37-38(自科6):830-837.
 WANG Yaming,ZHANG Jing,ZHENG Junbao.Research on 3D NonRigid Reconstruction Based on Acceleration Smoothness Constraints[J].Journal of Zhejiang Sci-Tech University,2017,37-38(自科6):830-837.
点击复制

基于加速度平滑约束的非刚体三维重建研究()
分享到:

浙江理工大学学报[ISSN:1673-3851/CN:33-1338/TS]

卷:
第37-38卷
期数:
2017年自科6期
页码:
830-837
栏目:
出版日期:
2017-11-10

文章信息/Info

Title:
Research on 3D NonRigid Reconstruction Based on Acceleration Smoothness Constraints
文章编号:
1673-3851 (2017) 06-0830-08
作者:
汪亚明张静郑俊褒
浙江理工大学信息学院,杭州 310018
Author(s):
WANG Yaming ZHANG Jing ZHENG Junbao
School of Information Science and Technology, Zhejiang Sci-Tech University, Hangzhou 310018, China
关键词:
非刚体三维重建加速度平滑约束解析解
分类号:
TP391.4
文献标志码:
A
摘要:
针对单一投影误差重建精度不高的问题,提出了基于加速度残差最小的轨迹平滑约束的非刚体三维重建方法。在运动轨迹连续性理论基础上,通过视频序列相邻帧的连续性进行分析,在三维重建投影误差基础上同时引入加速度平滑约束并建立目标函数,最后推导出闭式最优解析解。由于非刚性的运动复杂多变,提出的加速度连续性约束是对非刚体运动目标轨迹的物理特性的本质约束,更具有一般性和适应性。与四种典型的运动模型进行实验对比,结果表明该重建方法有效提高了重建的精确度,且实现简单、算法运算复杂度更小。

参考文献/References:

[1] ZHANG Y, GIBSON G M, HAY R, et al. A fast 3D reconstruction system with a lowcost camera accessory[J]. Scientific Reports,2015,5:10909.
[2] ALLDIECK T, KASSUBECK M, MAGNOR M. Optical   flowbased 3D human motion estimation from monocular video. (2017-03-01)[ 2017-06-27].
[3] BREGLER C, HERTZMANN A, BIEERMANN H. Recovering nonrigid 3D shape from image streams[C]// Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Los Alamitos. IEEE,2000:690-696.
[4] XIAO J, CHAI J, KANADE T. A closedform solution to nonrigid shape and motion recovery[J]. International Journal of Computer Vision,2004,67(2):573-587.
[5] GOTARDO P F U, MARTINEZ A M. Kernel nonrigid structure from motion[C]//Proceedings of the IEEE International Conference on Computer Vision. IEEE,2011:802-809.
[6] AKHTER I, SIMON T, KHAN S, et al. Bilinear spatiotemporal basis models[J]. ACM Transactions on Graphics,2012,31(2):1-12.
[7] WANG Y, YAN X, ZHENG J, et al. Sparse approximation for nonrigid structure from motion[J]. Journal of Robotics,2015(3):1-8.
[8] TAO L, MATUSZEWSKI B J. Robust deformable shape  reconstruction from monocular video with manifold forests[J]. Machine Vision and Applications,2016,27(6):1-19.
[9] TAO L, MATUSZEWSKI B J. Deformable shape reconstruction from monocular video with manifold forests[J]. Computer Analysis of Images and Patterns, Springer Berlin Heidelberg,2013:28-36.

相似文献/References:

[1]王晶,包晓敏,汪亚明.透视投影下基于轨迹基的非刚体三维结构恢复研究[J].浙江理工大学学报,2014,31-32(自科1):40.
 WANG Jing,BAO Xiao min,WANG Ya ming.Research on Recovery of Three dimensional Structure of Nonrigid Body  Based on Trajectory Basis under Perspective Projection[J].Journal of Zhejiang Sci-Tech University,2014,31-32(自科6):40.

备注/Memo

备注/Memo:
收稿日期: 2017-06-27
网络出版日期: 2017-10-10
基金项目: 浙江省自然科学基金重点项目(LZ15F020004);浙江理工大学521人才资助计划
作者简介: 张静(1991-),女,湖北麻城人,硕士研究生,主要从事图像处理、计算机视觉和模式识别等方面的研究
通信作者: 郑俊褒,E-mail:zhengjunbao@zstu.edu.cn
更新日期/Last Update: 2017-11-16