Shape Distributions of Nonlinear Dynamical Systems for Video-based Inference

TL;DR

This paper introduces a shape-based feature framework for analyzing nonlinear dynamical systems in videos, offering stable, data-driven representations that outperform traditional invariants in various activity recognition tasks.

Contribution

The paper proposes a novel shape distribution approach for dynamical analysis that is data-driven, stable across different time-series lengths, and applicable to video-based inference tasks.

Findings

Shape distributions effectively discriminate different dynamical behaviors.

The method remains stable with varying time-series lengths.

Experimental validation on motion capture and Kinect datasets supports the approach.

Abstract

This paper presents a shape-theoretic framework for dynamical analysis of nonlinear dynamical systems which appear frequently in several video-based inference tasks. Traditional approaches to dynamical modeling have included linear and nonlinear methods with their respective drawbacks. A novel approach we propose is the use of descriptors of the shape of the dynamical attractor as a feature representation of nature of dynamics. The proposed framework has two main advantages over traditional approaches: a) representation of the dynamical system is derived directly from the observational data, without any inherent assumptions, and b) the proposed features show stability under different time-series lengths where traditional dynamical invariants fail. We illustrate our idea using nonlinear dynamical models such as Lorenz and Rossler systems, where our feature representations (shape distribution) support our hypothesis that the local shape of the reconstructed phase space can be used as a discriminative feature. Our experimental analyses on these models also indicate that the proposed framework show stability for different time-series lengths, which is useful when the available number of samples are small/variable. The specific applications of interest in this paper are: 1) activity recognition using motion capture and RGBD sensors, 2) activity quality assessment for applications in stroke rehabilitation, and 3) dynamical scene classification. We provide experimental validation through action and gesture recognition experiments on motion capture and Kinect datasets. In all these scenarios, we show experimental evidence of the favorable properties of the proposed representation.