Parallelization strategies for markerless human motion capture

被引：4

作者：

Cano, Alberto ^{[1
]}

Yeguas-Bolivar, Enrique ^{[1
,2
]}

Munoz-Salinas, Rafael ^{[1
,2
]}

Medina-Carnicer, Rafael ^{[1
,2
]}

Ventura, Sebastian ^{[1
]}

机构：

[1] Univ Cordoba, Dept Comp Sci & Numer Anal, Cordoba, Spain

[2] Maimonides Inst Biomed Res IMIBIC, Cordoba, Spain

来源：

JOURNAL OF REAL-TIME IMAGE PROCESSING | 2018年 / 14卷 / 02期

关键词：

Markerless motion capture (MMOCAP); GPU; Tracking; 3D HUMAN MOTION; BODY MOTION; TRACKING; STEREOPHOTOGRAMMETRY; EVOLUTION;

D O I：

10.1007/s11554-014-0467-1

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Markerless motion capture (MMOCAP) is the problem of determining the pose of a person from images captured by one or several cameras simultaneously without using markers on the subject. Evaluation of the solutions is frequently the most time-consuming task, making most of the proposed methods inapplicable in real-time scenarios. This paper presents an efficient approach to parallelize the evaluation of the solutions in CPUs and GPUs. Our proposal is experimentally compared on six sequences of the HumanEva-I dataset using the CMAES algorithm. Multiple algorithm's configurations were tested to analyze the best trade-off with regard to the accuracy and computing time. The proposed methods obtain speedups of 8 in multi-core CPUs, 30 in a single GPU and up to 110 using 4 GPUs.

引用

页码：453 / 467

页数：15

共 49 条

[41] Model-driven segmentation of articulating humans in Laplacian Eigenspace [J].

Sundaresan, Aravind ;

Chellappa, Rama .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2008, 30 (10) :1771-1785

[42]

Ugo D. C, 2005, GAIT POSTURE, V21, P226

[43] Particle Swarm Optimization and Differential Evolution for model-based object detection [J].

Ugolotti, Roberto ;

Nashed, Youssef S. G. ;

Mesejo, Pablo ;

Ivekovic, Spela ;

Mussi, Luca ;

Cagnoni, Stefano .

APPLIED SOFT COMPUTING, 2013, 13 (06) :3092-3105

[44] Comparing evolutionary algorithms and particle filters for Markerless Human Motion Capture [J].

Yeguas-Bolivar, Enrique ;

Munoz-Salinas, Rafael ;

Medina-Carnicer, Rafael ;

Carmona-Poyato, Angel .

APPLIED SOFT COMPUTING, 2014, 17 :153-166

[45] CUDA ACCELERATION OF 3D DYNAMIC SCENE RECONSTRUCTION AND 3D MOTION ESTIMATION FOR MOTION CAPTURE [J].

Zhang, Zheng ;

Seah, Hock Soon .

PROCEEDINGS OF THE 2012 IEEE 18TH INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED SYSTEMS (ICPADS 2012), 2012, :284-291

[46] GPU-Accelerated Real-Time Tracking of Full-Body Motion With Multi-Layer Search [J].

Zhang, Zheng ;

Seah, Hock Soon ;

Quah, Chee Kwang ;

Sun, Jixiang .

IEEE TRANSACTIONS ON MULTIMEDIA, 2013, 15 (01) :106-119

[47] Generative tracking of 3D human motion by hierarchical annealed genetic algorithm [J].

Zhao, Xu ;

Liu, Yuncai .

PATTERN RECOGNITION, 2008, 41 (08) :2470-2483

[48] Human motion tracking for rehabilitation-A survey [J].

Zhou, Huiyu ;

Hu, Huosheng .

BIOMEDICAL SIGNAL PROCESSING AND CONTROL, 2008, 3 (01) :1-18

[49] GPU-based Parallel Particle Swarm Optimization [J].

Zhou, You ;

Tan, Ying .

2009 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION, VOLS 1-5, 2009, :1493-+

← 1 2 3 4 5 →