Coupling Gaussian Process Dynamical Models with Product-of-Experts Kernels

Velychko, Dmytro; Endres, Dominik; Taubert, Nick; Giese, Martin A.

doi:10.1007/978-3-319-11179-7_76

Cited by 6 publications

(10 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We therefore plan to combine the advantages of modular MPs with those of dynamical MPs, where (in contrast to the common use of DMPs) we also include the dynamics (in a mechanical sense) of the whole body system. Our first steps in this direction are promising [18,48], but a demonstration of these modular dynamical MPs on large Figure 13: Generated walking sequence with 20+3 steps of varying physical step length (from 150mm to 400mm back to 150mm), validated in virtual robot OpenHRP. The screen shots are captured with a frequency of 1Hz.…”

Section: Discussionmentioning

confidence: 99%

“…Also, it was demonstrated in [16] that individual DMPs can encode both transient and rhythmic components of a movement. However, they are not (yet) modular (but see [17,18]), i.e. one DMP drives all relevant degrees of freedom.…”

Section: Movement Primitivesmentioning

confidence: 99%

“…For all others, it can be obtained from eqn. (18) by substituting X k,i,s with the end of the previous movement, and the posterior parameters of the training movements with the posterior parameters of the new movement.…”

Section: Generating New Movementsmentioning

confidence: 99%

See 2 more Smart Citations

A novel approach for the generation of complex humanoid walking sequences based on a combination of optimal control and learning of movement primitives

Clever

Harant

Koch

et al. 2016

Robotics and Autonomous Systems

Self Cite

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Section: Movement Primitivesmentioning

confidence: 99%

See 1 more Smart Citation

A novel approach for the generation of complex humanoid walking sequences based on a combination of optimal control and learning of movement primitives

Clever

Harant

Koch

et al. 2016

Robotics and Autonomous Systems

Self Cite

View full text Add to dashboard Cite

“…This is a special case of the variational coupled GPDM described in 3.1.5. (Velychko et al 2014). The cGPDM was proposed to make GPDMs modular.…”

Section: Gaussian Process Dynamical Model (Gpdm)mentioning

confidence: 99%

Untitled

2019

ACM Trans. Appl. Percept.

View full text Add to dashboard Cite

We compared the perceptual validity of human avatar walking animations driven by six different representations of human movement using a graphics Turing test. All six representations are based on movement primitives (MPs), which are predictive models of full-body movement that differ in their complexity and prediction mechanism. Assuming that humans are experts at perceiving biological movement from noisy sensory signals, it follows that these percepts should be describable by a suitably constructed Bayesian ideal observer model. We build such models from MPs and investigate if the perceived naturalness of human animations are predictable from approximate Bayesian model scores of the MPs. We found that certain MP-based representations are capable of producing movements that are perceptually indistinguishable from natural movements. Furthermore, approximate Bayesian model scores of these representations can be used to predict perceived naturalness. In particular, we could show that movement dynamics are more important for perceived naturalness of human animations than single frame poses. This indicates that perception of human animations is highly sensitive to their temporal coherence. More generally, our results add evidence for a shared MP-representation of action and perception. Even though the motivation of our work is primarily drawn from neuroscience, we expect that our results will be applicable in virtual and augmented reality settings, when perceptually plausible human avatar movements are required.

show abstract

“…We describe a model that learns MPs composed of coupled dynamical systems and associated kinematics mappings, where both components are learned, thus lifting the DMP’s restriction of canonical dynamics. We build on the Coupled Gaussian Process Dynamical Model (CGPDM) by [ 14 ], which combines the advantages of modularity and flexibility in the dynamics, at least theoretically. In a CGPDM, the temporal evolution functions for the latent dynamical systems are drawn out of a Gaussian process (GP) prior [ 15 ].…”

Section: Introductionmentioning

confidence: 99%

Making the Coupled Gaussian Process Dynamical Model Modular and Scalable with Variational Approximations

Velychko

Knopp

Endres

2018

Entropy

Self Cite

View full text Add to dashboard Cite

We describe a sparse, variational posterior approximation to the Coupled Gaussian Process Dynamical Model (CGPDM), which is a latent space coupled dynamical model in discrete time. The purpose of the approximation is threefold: first, to reduce training time of the model; second, to enable modular re-use of learned dynamics; and, third, to store these learned dynamics compactly. Our target applications here are human movement primitive (MP) models, where an MP is a reusable spatiotemporal component, or "module" of a human full-body movement. Besides re-usability of learned MPs, compactness is crucial, to allow for the storage of a large library of movements. We first derive the variational approximation, illustrate it on toy data, test its predictions against a range of other MP models and finally compare movements produced by the model against human perceptual expectations. We show that the variational CGPDM outperforms several other MP models on movement trajectory prediction. Furthermore, human observers find its movements nearly indistinguishable from replays of natural movement recordings for a very compact parameterization of the approximation.

show abstract

Coupling Gaussian Process Dynamical Models with Product-of-Experts Kernels

Cited by 6 publications

References 14 publications

A novel approach for the generation of complex humanoid walking sequences based on a combination of optimal control and learning of movement primitives

A novel approach for the generation of complex humanoid walking sequences based on a combination of optimal control and learning of movement primitives

Untitled

Making the Coupled Gaussian Process Dynamical Model Modular and Scalable with Variational Approximations

Contact Info

Product

Resources

About