An Eigen-based motion retrieval method for real-time animation

Wang, Pengjie; Lau, Rynson W. H.; Pan, Zhigeng; Wang, Jiang; Song, Haiyu

doi:10.1016/j.cag.2013.11.008

Cited by 12 publications

(8 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2 For simplicity and readability, some abbreviated concepts that appear repeatedly in the following are stated in Table 1 intuitively.…”

Section: Mocap Data Representationmentioning

confidence: 99%

See 1 more Smart Citation

Hierarchical block-based incomplete human mocap data recovery using adaptive nonnegative matrix factorization

Peng

Liu

et al. 2015

Computers & Graphics

View full text Add to dashboard Cite

“…2 For simplicity and readability, some abbreviated concepts that appear repeatedly in the following are stated in Table 1 intuitively.…”

Section: Mocap Data Representationmentioning

confidence: 99%

“…In recent years, different kinds of human mocap data have been extensively utilized in computer animation [2,3], sports training [4], human computer interactions, and other virtual reality applications [5,6]. For instance, the impressive Avatar film has shown the great success of motion capture techniques involved in human animation [7].…”

Section: Introductionmentioning

confidence: 99%

Hierarchical block-based incomplete human mocap data recovery using adaptive nonnegative matrix factorization

Peng

Liu

et al. 2015

Computers & Graphics

View full text Add to dashboard Cite

“…Lv et al [25] measure the distance between motion signatures by marker-based geometry, which shows outstanding retrieval performance on various datasets. Wang et al [38] represent body parts by eigenvectors and match the individual and integrated similarities based on the body segments.…”

Section: Related Workmentioning

confidence: 99%

“…Based on common sense, human skeleton can be roughly divided into five main parts of four limbs and a trunk [9,38]. An example body part segmentation of 24 joints is given in Fig.…”

Section: Bpr Descriptorsmentioning

confidence: 99%

Retrieval of spatial–temporal motion topics from 3D skeleton data

Men

Leung

2019

Vis Comput

View full text Add to dashboard Cite

Retrieval of a specific human motion from 3D skeleton data is intractable because of its articulated complexity. We propose a context-based motion document formation method to reflect geometric variations by calculating covariance descriptors among skeletal joint locations and joint relative distances, and temporal variations by performing a coarse-to-fine segmentation on the motion sequence. The descriptors of query motion traverse all the motion categories to lock its motion words, which can be regarded as the basic units of a motion document. The discrete motion words of different spatiotemporal descriptors are also mapped to divergent index ranges to add prior knowledge of motion with temporal order to latent Dirichlet allocation (LDA). The similarity matching is based on motion-topic distributions from LDA with semantic meanings. The experiments on public datasets show the effectiveness and robustness of the proposed method over existing models.

show abstract

“…Along this way, Li et al [7] utilised the singular value decomposition (SVD) on multi-attribute motion matrices and obtained one representative vector for motion indexing, while Pradhan et al [8] applied the SVD on multidimensional 3D mocap data and indexed the sub-body motion for whole action retrieval. Similarly, Barbic et al [9] employed the principal component analysis (PCA) to detect the inherent motion changes, while Wang et al [10] selected an eigen-based vector to characterise each motion clip. Although these dimension reduction methods are able to capture the significant motion variances, the non-linear information within the articulated complexities is not sufficiently exploited.…”

Section: Related Workmentioning

confidence: 99%

Efficient human motion capture data annotation via multi‐view spatiotemporal feature fusion

Liu

Meng²,

Peng

et al. 2018

IET signal process.

View full text Add to dashboard Cite

The availability of large motion capture (mocap) data has sparked a great motivation for computer animation, and the task of automatically annotating complex mocap sequences plays an important role in the efficient motion analysis. To this end, this study presents an efficient human mocap data annotation approach by using multi-view spatiotemporal feature fusion. First, the authors exploit an improved hierarchical aligned cluster analysis algorithm to divide the unknown human mocap sequence into several sub-motion clips, and each sub-motion clip incorporates a particular semantic meaning. Then, the two kinds of multi-view features, namely most informative central distances and most informative geometric angles, are discriminatively extracted and temporally modelled by a Fourier temporal pyramid to complementarily characterise each motion clip. Finally, the authors utilise the discriminant correlation analysis to fuse these two types of motion features and further employ an extreme learning machine to annotate each sub-motion clip. The extensive experiments tested on the public available database have demonstrated the effectiveness of the proposed approach in comparison with the existing counterparts.

show abstract

An Eigen-based motion retrieval method for real-time animation

Cited by 12 publications

References 44 publications

Hierarchical block-based incomplete human mocap data recovery using adaptive nonnegative matrix factorization

Hierarchical block-based incomplete human mocap data recovery using adaptive nonnegative matrix factorization

Retrieval of spatial–temporal motion topics from 3D skeleton data

Efficient human motion capture data annotation via multi‐view spatiotemporal feature fusion

Contact Info

Product

Resources

About