LEARNING NONLINEAR MANIFOLDS OF DYNAMIC TEXTURES

Ishan Awasthi, Ahmed Elgammal

2006

Abstract

Dynamic textures are sequences of images of moving scenes that show stationarity properties in time. Eg: waves, flame, fountain, etc. Recent attempts at generating, potentially, infinitely long sequences model the dynamic texture as a Linear Dynamic System. This assumes a linear correlation in the input sequence. Most real world sequences however, exhibit nonlinear correlation between frames. In this paper, we propose a technique of generating dynamic textures using a low dimension model that preserves the non-linear correlation. We use nonlinear dimensionality reduction to create an embedding of the input sequence. Using this embedding, a nonlinear mapping is learnt from the embedded space into the image input space. Any input is represented by a linear combination of nonlinear bases functions centered along the manifold in the embedded space. A spline is used to move along the input manifold in this embedded space as a similar manifold is created for the output. The nonlinear mapping learnt on the input is used to map this new manifold into a sequence in the image space. Output sequences, thus created, contain images never present in the original sequence and are very realistic.

Paper Citation

in Harvard Style

Awasthi I. and Elgammal A. (2006). LEARNING NONLINEAR MANIFOLDS OF DYNAMIC TEXTURES . In Proceedings of the First International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP, ISBN 972-8865-40-6, pages 243-250. DOI: 10.5220/0001378202430250

in Bibtex Style

@conference{visapp06,
author={Ishan Awasthi and Ahmed Elgammal},
title={LEARNING NONLINEAR MANIFOLDS OF DYNAMIC TEXTURES},
booktitle={Proceedings of the First International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP,},
year={2006},
pages={243-250},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0001378202430250},
isbn={972-8865-40-6},
}

in EndNote Style

TY - CONF
JO - Proceedings of the First International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP,
TI - LEARNING NONLINEAR MANIFOLDS OF DYNAMIC TEXTURES
SN - 972-8865-40-6
AU - Awasthi I.
AU - Elgammal A.
PY - 2006
SP - 243
EP - 250
DO - 10.5220/0001378202430250