• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利：CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法，用中文像聊天一样搜索，搜遍4000万医学文献。AI智能推荐，让科研检索更轻松。

立即免费搜索

文件翻译

保留排版，准确专业，支持PDF/Word/PPT等文件格式，支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述，25分钟生成高质量综述，智能提取关键信息，辅助科研写作。

立即免费体验

作为用于姿态估计的4D-DPM模型约束的对偶四元数

Dual Quaternions as Constraints in 4D-DPM Models for Pose Estimation.

作者信息

Martinez-Berti Enrique, Sánchez-Salmerón Antonio-José, Ricolfe-Viala Carlos

机构信息

Departamento de Ingeniería de Sistemas y Automática, Instituto de Automática e informática Industrial, Universitat Politècnica de València, València 46022, Spain.

出版信息

Sensors (Basel). 2017 Aug 19;17(8):1913. doi: 10.3390/s17081913.

DOI:10.3390/s17081913
PMID:28825627
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC5579524/
Abstract

The goal of this research work is to improve the accuracy of human pose estimation using the Deformation Part Model (DPM) without increasing computational complexity. First, the proposed method seeks to improve pose estimation accuracy by adding the depth channel to DPM, which was formerly defined based only on red-green-blue (RGB) channels, in order to obtain a four-dimensional DPM (4D-DPM). In addition, computational complexity can be controlled by reducing the number of joints by taking it into account in a reduced 4D-DPM. Finally, complete solutions are obtained by solving the omitted joints by using inverse kinematics models. In this context, the main goal of this paper is to analyze the effect on pose estimation timing cost when using dual quaternions to solve the inverse kinematics.

摘要

这项研究工作的目标是在不增加计算复杂度的情况下，使用变形部件模型（DPM）提高人体姿态估计的准确性。首先，所提出的方法旨在通过将深度通道添加到以前仅基于红-绿-蓝（RGB）通道定义的DPM中，来提高姿态估计精度，以获得四维DPM（4D-DPM）。此外，通过在简化的4D-DPM中考虑减少关节数量，可以控制计算复杂度。最后，通过使用逆运动学模型求解遗漏的关节来获得完整的解决方案。在此背景下，本文的主要目标是分析使用对偶四元数求解逆运动学时对姿态估计时间成本的影响。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/eab6/5579524/a734c515d2bf/sensors-17-01913-g008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/eab6/5579524/5c7744ec697a/sensors-17-01913-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/eab6/5579524/1d3a0907663c/sensors-17-01913-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/eab6/5579524/bbdf5f042c04/sensors-17-01913-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/eab6/5579524/c9cecf3ce2a9/sensors-17-01913-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/eab6/5579524/7db27f68fe7c/sensors-17-01913-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/eab6/5579524/78fd08588195/sensors-17-01913-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/eab6/5579524/3eda6bd674fa/sensors-17-01913-g007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/eab6/5579524/a734c515d2bf/sensors-17-01913-g008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/eab6/5579524/5c7744ec697a/sensors-17-01913-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/eab6/5579524/1d3a0907663c/sensors-17-01913-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/eab6/5579524/bbdf5f042c04/sensors-17-01913-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/eab6/5579524/c9cecf3ce2a9/sensors-17-01913-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/eab6/5579524/7db27f68fe7c/sensors-17-01913-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/eab6/5579524/78fd08588195/sensors-17-01913-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/eab6/5579524/3eda6bd674fa/sensors-17-01913-g007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/eab6/5579524/a734c515d2bf/sensors-17-01913-g008.jpg

相似文献

1
Dual Quaternions as Constraints in 4D-DPM Models for Pose Estimation.作为用于姿态估计的4D-DPM模型约束的对偶四元数
Sensors (Basel). 2017 Aug 19;17(8):1913. doi: 10.3390/s17081913.
2
Simultaneous Recognition and Relative Pose Estimation of 3D Objects Using 4D Orthonormal Moments.使用四维正交矩的三维物体同时识别与相对姿态估计
Sensors (Basel). 2017 Sep 15;17(9):2122. doi: 10.3390/s17092122.
3
Interpolation of three dimensional kinematics with dual-quaternions.使用对偶四元数对三维运动学进行插值
J Biomech. 2017 Jan 25;51:105-110. doi: 10.1016/j.jbiomech.2016.10.028. Epub 2016 Oct 27.
4
Inertial measurement unit-based pose estimation: Analyzing and reducing sensitivity to sensor placement and body measures.基于惯性测量单元的姿态估计：分析并降低对传感器放置和身体尺寸的敏感度。
J Rehabil Assist Technol Eng. 2019 Jan 14;6:2055668318813455. doi: 10.1177/2055668318813455. eCollection 2019 Jan-Dec.
5
U-net-based deformation vector field estimation for motion-compensated 4D-CBCT reconstruction.基于U-net的形变矢量场估计用于运动补偿4D-CBCT重建。
Med Phys. 2020 Jul;47(7):3000-3012. doi: 10.1002/mp.14150. Epub 2020 Apr 27.
6
Highly Accurate and Fully Automatic 3D Head Pose Estimation and Eye Gaze Estimation Using RGB-3D Sensors and 3D Morphable Models.基于 RGB-3D 传感器和 3D 可变形模型的高精度全自动 3D 头部姿态估计和眼动估计。
Sensors (Basel). 2018 Dec 5;18(12):4280. doi: 10.3390/s18124280.
7
Model-based extended quaternion Kalman filter to inertial orientation tracking of arbitrary kinematic chains.基于模型的扩展四元数卡尔曼滤波器用于任意运动链的惯性方向跟踪。
Springerplus. 2016 Nov 14;5(1):1965. doi: 10.1186/s40064-016-3653-8. eCollection 2016.
8
Dynamic Pose Estimation Using Multiple RGB-D Cameras.基于多台 RGB-D 相机的动态姿态估计
Sensors (Basel). 2018 Nov 10;18(11):3865. doi: 10.3390/s18113865.
9
Kalman Filtering for Attitude Estimation with Quaternions and Concepts from Manifold Theory.四元数和流形理论概念的姿态估计的卡尔曼滤波。
Sensors (Basel). 2019 Jan 3;19(1):149. doi: 10.3390/s19010149.
10
A Kalman Filter for Nonlinear Attitude Estimation Using Time Variable Matrices and Quaternions.一种使用时变矩阵和四元数的非线性姿态估计卡尔曼滤波器。
Sensors (Basel). 2020 Nov 25;20(23):6731. doi: 10.3390/s20236731.

引用本文的文献

1
Dual-Quaternion Analytic LQR Control Design for Spacecraft Proximity Operations.用于航天器近距离操作的对偶四元数解析线性二次调节器控制设计
Sensors (Basel). 2021 May 21;21(11):3597. doi: 10.3390/s21113597.

本文引用的文献

1
Real-Time Simultaneous Pose and Shape Estimation for Articulated Objects Using a Single Depth Camera.使用单目深度相机实时估计关节物体的姿态和形状。
IEEE Trans Pattern Anal Mach Intell. 2016 Aug;38(8):1517-32. doi: 10.1109/TPAMI.2016.2557783. Epub 2016 Apr 21.
2
Articulated and Generalized Gaussian Kernel Correlation for Human Pose Estimation.关节和广义高斯核相关的人体姿态估计
IEEE Trans Image Process. 2016 Feb;25(2):776-89. doi: 10.1109/TIP.2015.2507445. Epub 2015 Dec 9.
3
Learning Actionlet Ensemble for 3D Human Action Recognition.
学习动作单元集以进行 3D 人体动作识别。
IEEE Trans Pattern Anal Mach Intell. 2014 May;36(5):914-27. doi: 10.1109/TPAMI.2013.198.
4
Articulated human detection with flexible mixtures of parts.具有灵活部件混合的关节式人体检测。
IEEE Trans Pattern Anal Mach Intell. 2013 Dec;35(12):2878-90. doi: 10.1109/TPAMI.2012.261.
5
Efficient human pose estimation from single depth images.基于单目深度图像的高效人体姿态估计。
IEEE Trans Pattern Anal Mach Intell. 2013 Dec;35(12):2821-40. doi: 10.1109/TPAMI.2012.241.
6
Accurate calibration with highly distorted images.使用高度失真的图像进行精确校准。
Appl Opt. 2012 Jan 1;51(1):89-101. doi: 10.1364/AO.51.000089.
7
Calibration of a wide angle stereoscopic system.广角立体视镜系统校准。
Opt Lett. 2011 Aug 15;36(16):3064-6. doi: 10.1364/OL.36.003064.