Style subspaces for character animation

Liu, Gengdai; Pan, Zhigeng; Lin, Zuoyan

doi:10.1002/cav.254

Cited by 9 publications

(8 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this category, we can name methods that utilize independent component analysis (ICA) [19,20] and weighted principal component analysis (wPCA) [21]. Probabilistic methods [22] and Markov models [23][24][25] have also been utilized for aligning time series.…”

Section: Related Workmentioning

confidence: 99%

Correlation-optimized time warping for motion

Etemad

Arya

2014

Vis Comput

View full text Add to dashboard Cite

Retrieval and comparative editing/modeling of motion data require temporal alignment. In other words, for such processes to perform accurately, critical features of motion sequences need to occur simultaneously. In this paper, we propose correlation-optimized time warping (CoTW) for aligning motion data. CoTW utilizes a correlation-based objective function for characterizing alignment. The method solves an optimization problem to determine the optimum warping degree for different segments of the sequence. Using segment-wise interpolated warping, smooth motion trajectories are achieved that can be readily used for animation. Our method allows for manual tuning of the parameters, resulting in high customizability with respect to the number of actions in a single sequence as well as spatial regions of interest within the character model. Moreover, measures are taken to reduce distortion caused by over-warping. The framework also allows for automatic selection of an optimum reference when multiple sequences are available. Experimental results demonstrate the very accurate performance of CoTW compared to other techniques such as dynamic time warping, derivative dynamic time warping and canonical time warping. The mentioned customization capabilities are also illustrated. Electronic supplementary materialThe online version of this article (

show abstract

Section: Related Workmentioning

confidence: 99%

Correlation-optimized time warping for motion

Etemad

Arya

2014

Vis Comput

View full text Add to dashboard Cite

show abstract

“…On the basis of this work, Liu et al . adopt independent feature subspace analysis (IFSA) , which is a modification of ICA, to extract styles. Different from ICA, IFSA considers some high‐order invariant features that describe the style more precisely.…”

Section: Related Workmentioning

confidence: 99%

“…Most researchers use dynamic time warping method to align the motions to a reference one. However, the aligned results may lead to jerkiness, high discontinuities or long steady postures because the warp path contains many consecutive horizontal or vertical steps .…”

Section: Motion Analysis Processmentioning

confidence: 99%

Flexible editing of human motion by three‐way decomposition

Liang

Wang

et al. 2013

Computer Animation & Virtual

View full text Add to dashboard Cite

This paper proposes a new generative model for flexible editing of human motion. Different from previous work, three intuitive factors of motion, namely, content, identity and style, can be manipulated directly with the new model. With the new generative model, motion editing can be achieved in various aspects, including transferring an unknown style from an actor to another, synthesizing other styles for an unknown actor and generating a new motion with other content. Copyright © 2013 John Wiley & Sons, Ltd.

show abstract

“…Synthesis, which is the second category, can be used to generate motion sequences, or alter existing ones, to achieve desired functional and perceptual properties. Finally, extraction of SFs is the third possible route [21]. Extraction of features is often of little practical significance alone.…”

Section: List Of Abbreviationsmentioning

confidence: 99%

“…On the other hand, in many cases, processing human motion requires motion data to be perfectly aligned. For example, it has been demonstrated that altering [90], blending [23], extraction [21], and translation [24] of motion require temporally Interpretation Synthesis Motion Sequence Information aligned data. Consequently, aligning motion is a critical step for both interpretation and synthesis procedures.…”

Section: Human Motion: Perceptionmentioning

confidence: 99%

Perceptually Guided Processing of Style and Affect in Human Motion for Multimedia Applications

Etemad¹

View full text Add to dashboard Cite

Computer graphics and animation, as a direct result of advancements in hardware and software, have become broad and demanding areas of research. Animation of human motion is a major component in the field that has attracted many due to its significance in movies, games, and virtual environments. We propose that processing features for style and affect, which are fundamental determinants of personality and naturally appearing motion, should be carried out through perceptually guided processing techniques. In this dissertation, we employ this approach and develop a set of tools for extraction, synthesis, and analysis of affective and stylistic motion features.Temporal alignment is one of the most common issues in processing motion data. Accordingly, we first propose a new time warping technique for motion. The proposed method outperforms several existing techniques and has advantages such as precise alignment, low distortion, smooth warped motion trajectories, and high customizability.Many motion processing techniques utilize incremental (joint-to-joint) processing of motion sequences. In addition, some systems process only selected joints or regions of the body. Hence, it is imperative to verify whether partial or subsets of computational solutions can lead to perceptually accurate results. Accordingly, we investigate and validate the notion of additivity in perception of affect from motion.A system capable of extracting style/affect features from motion data using spline optimization is then introduced. Our method has several advantages over existing techniques, namely extracting the features as three separate movement, posture, and time components, which are the perceptual and functional sources for stylistic/affective motion. Our method also performs in Cartesian or joint-angle spaces rather than ii Eigen/latent sub-spaces which is the common trend in existing techniques.Towards synthesis of style/affect features, a perception-based expert-driven approach is used. Gaussian radial basis functions (RBFs) are first introduced as mathematical constructs for stylistic/affective features. A user interface is then developed using which animators can utilize these basis functions to synthesize the desired stylistic/affective features. Through analysis and in depth study of data collected from several animators, expert-driven perceptual shortcuts for generation of different stylistic/affective themes are derived. The features also shed light on various aspects of execution and perception of style/affect. A unified system capable of both classification and translation of stylistic/affective features in motion is subsequently developed using ensembles of Gaussian RBF neural networks. The recognition module of the system outperforms several other classifiers and the style translation module produces results that are validated as perceptually accurate by viewers. Finally, to provide a set of guidelines for animators, based on which stylistic/affective features can be modified or added to motion data, an empirical paradigm i...

show abstract

Style subspaces for character animation

Cited by 9 publications

References 15 publications

Correlation-optimized time warping for motion

Correlation-optimized time warping for motion

Flexible editing of human motion by three‐way decomposition

Perceptually Guided Processing of Style and Affect in Human Motion for Multimedia Applications

Contact Info

Product

Resources

About