Decomposition and dictionary learning for 3D trajectories

Abstract

A new model for describing a three-dimensional (3D) trajectory is proposed in this article. The studied trajectory is viewed as a linear combination of rotatable 3D patterns. The resulting model is thus 3D rotation invariant (3DRI). Moreover, the temporal patterns are considered as shift-invariant. This article is divided into two parts based on this model. On the one hand, the 3DRI decomposition estimates the active patterns, their coefficients, their rotations and their shift parameters. Based on sparse approximation, this is carried out by two non-convex optimizations: 3DRI matching pursuit (3DRI-MP) and 3DRI orthogonal matching pursuit (3DRI-OMP). On the other hand, a 3DRI learning method learns the characteristic patterns of a database through a 3DRI dictionary learning algorithm (3DRI-DLA). The proposed algorithms are first applied to simulation data to evaluate their performances and to compare them to other algorithms. Then, they are applied to real motion data of cued speech, to learn the 3D trajectory patterns characteristic of this gestural language.

Color Sparse Representations for Image Processing: Review, Models, and Prospects

“…Among the possibilities, the normalization such that D m W m F = D m F = 1 keeps the atom norm equal after linear transformation. This normalization appears to be an appropriate choice, as it has been used efficiently for matrices W m restricted to rotations [51].…”

Sparse Approximation for Nonrigid Structure from Motion

“…So this paper introduces a new method which uses sparse approximation method to represent the trajectory curves instead of traditional trajectory bases method. And the sparse approximation provides a class of algorithms that learn basis functions only when they capture higher-level features in the input data [14]. Moreover, an overcomplete atom dictionary will be used in this method other than trajectory bases.…”

Spatial and rotation invariant 3D gesture recognition based on sparse representation

“…Sparse representation is convenient to handle the sparsity inherent to human motion which can be spatially invariant (different actions can share a similar movement) and temporally invariant (the same action can be performed at different times) [4]. Human motion can be recorded as time series of position, orientation, speed or acceleration data [13] from one or multiple joints of the human body.…”

“…The proposed algorithm is able to extract the most representative motion patterns from a given set of gestures and uses this information to enable the real-time classification of 3D gestures. Our approach, inspired by state of the art 3D sparse representation algorithms [4], provides a robust gesture classifier, that is tolerant to the speed, the scale and the rotation of the gesture. Furthermore, the sparse representation of human motion is robust to noise (only salient features are kept) and generates a compact representation.…”