Extraction of key postures from 3D human motion data for choreography summarization

被引:0
作者
Rallis, Ioannis [1 ]
Georgoulas, Ioannis [1 ]
Doulamis, Nikolaos [1 ]
Voulodimos, Athanasios [1 ]
Terzopoulos, Panagiotis [2 ]
机构
[1] Natl Tech Univ Athens, Athens, Greece
[2] Metis Balt, Vilnius, Lithuania
来源
2017 9TH INTERNATIONAL CONFERENCE ON VIRTUAL WORLDS AND GAMES FOR SERIOUS APPLICATIONS (VS-GAMES) | 2017年
关键词
motion capturing; choreogrpahy summarization; clustering; k-means; key posture extraction; CAPTURE;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Modelling and digitizing performing arts through motion capturing interfaces is an important aspect for the analysis, processing and documentation of intangible cultural heritage assets. However, existing modelling approaches may involve huge amounts of information which are difficult to process, store and analyze. To address these limitations, usually a skeleton describing the dancer motion is extracted. However, often the complexity still remains due to the high spatio-temporal dependencies of the detected skeleton joints. In this paper, an alternative approach is presented: choreography summarization. This means that a very small number of image frames are extracted to represent a choreography, thus significantly reducing the processing and storage complexity. In our approach the problem of choreography summarization is treated as an unsupervised clustering approach. Evaluation indices are introduced for monitoring the summarization performance. Experimental results on real-life dancing performances verifies the capability of the proposed method to capture the main patterns of a choreography with a very small number of trajectory points.
引用
收藏
页码:94 / 101
页数:8
相关论文
共 50 条
[41]   A SMOOTH 6DOF MOTION PRIOR FOR EFFICIENT 3D SURFACE TRACKING [J].
Popham, Thomas ;
Wilson, Roland ;
Bhalerao, Abhir .
2010 3DTV-CONFERENCE: THE TRUE VISION - CAPTURE, TRANSMISSION AND DISPLAY OF 3D VIDEO (3DTV-CON 2010), 2010,
[42]   Analysis of anticipation by 3D motion capturing - a new method presented in karate kumite [J].
Petri, Katharina ;
Lichtenstein, Marvin ;
Bandow, Nicole ;
Campe, Sebastian ;
Wechselberger, Marcus ;
Sprenger, Dominik ;
Kaczmarek, Felix ;
Emmermacher, Peter ;
Witte, Kerstin .
JOURNAL OF SPORTS SCIENCES, 2017, 35 (02) :130-135
[43]   3D hypothesis clustering for cross-view matching in multiperson motion capture [J].
Miaopeng Li ;
Zimeng Zhou ;
Xinguo Liu .
Computational Visual Media, 2020, 6 (02) :147-156
[44]   Swarm Intelligence Based Searching Schemes for Articulated 3D Body Motion Tracking [J].
Kwolek, Bogdan ;
Krzeszowski, Tomasz ;
Wojciechowski, Konrad .
ADVANCED CONCEPTS FOR INTELLIGENT VISION SYSTEMS, 2011, 6915 :115-126
[45]   Reconstruction of 3D human motion in real-time using particle swarm optimization with GPU-accelerated fitness function [J].
Kwolek, Bogdan ;
Rymut, Boguslaw .
JOURNAL OF REAL-TIME IMAGE PROCESSING, 2020, 17 (04) :821-838
[46]   Cluster analysis of 3D seismic data for oil and gas exploration [J].
Moraes, D. R. S. ;
Espindola, R. P. ;
Evsukoff, A. G. ;
Ebecken, N. F. F. .
DATA MINING VII: DATA, TEXT AND WEB MINING AND THEIR BUSINESS APPLICATIONS, 2006, 37 :63-+
[47]   Resolution Influence on 3D Anthropometric Data Clustering for Fitting Design [J].
Niu, Jianwei ;
Li, Zhizhong ;
Salvendy, Gavriel .
INDUSTRIAL HEALTH, 2009, 47 (05) :578-585
[48]   Local adaptive parameterization for the history matching of 3D seismic data [J].
Sebastien Da Veiga ;
Veronique Gervais .
Computational Geosciences, 2012, 16 :483-498
[49]   A data-centric unsupervised 3D mesh segmentation method [J].
Sivri, Talya Tumer ;
Sahillioglu, Yusuf .
VISUAL COMPUTER, 2024, 40 (04) :2237-2249
[50]   A data-centric unsupervised 3D mesh segmentation method [J].
Talya Tümer Sivri ;
Yusuf Sahillioğlu .
The Visual Computer, 2024, 40 :2237-2249