Does Dimensionality Reduction improve the Quality of Motion Interpolation?
Proc. 17th European Symposium on Artificial Neural Networks (ESANN ’09)
Date
2009Author
Bitzer, Sebastian
Klanke, Stefan
Vijayakumar, Sethu
Metadata
Abstract
In recent years nonlinear dimensionality reduction has frequently
been suggested for the modelling of high-dimensional motion data.
While it is intuitively plausible to use dimensionality reduction to recover
low dimensional manifolds which compactly represent a given set of movements,
there is a lack of critical investigation into the quality of resulting
representations, in particular with respect to generalisability. Furthermore
it is unclear how consistently particular methods can achieve good results.
Here we use a set of robotic motion data for which we know the ground
truth to evaluate a range of nonlinear dimensionality reduction methods
with respect to the quality of motion interpolation. We show that results
are extremely sensitive to parameter settings and data set used, but
that dimensionality reduction can potentially improve the quality of linear motion interpolation, in particular in the presence of noise.