Behavior Classification By Eigendecomposition of Periodic Motions
.
Behavior classification by eigendecomposition of periodic motions.
Pattern Recognition, 38(8):1033-1043, 2005
Online Version
A pdf version is available for download.
Abstract
We show how periodic motions can be represented by a small number of eigenshapes that capture the whole dynamic mechanism of the motion. Spectral decomposition of a silhouette of a moving object serves as a basis for behavior classification by principle component analysis. The boundary contour of the walking dog, for example, is first computed efficiently and accurately. After normalization, the implicit representation of a sequence of silhouette contours given by their corresponding binary images, is used for generating eigenshapes for the given motion. Singular value decomposition produces these eigenshapes that are then used to analyze the sequence. We show examples of object as well as behavior classification based on the eigendecomposition of the binary silhouette sequence.
Keywords
- Visual Motion,
- Segmentation And Tracking,
- Object Recognition,
- Activity Recognition,
- Non-rigid Motion,
- Active Contours,
- Periodicity Analysis,
- Motion-based Classification,
- Eigenshapes.
Co-authors
Bibtex Entry
@article{GoldenbergKRR05a,
title = {Behavior classification by eigendecomposition of periodic motions},
author = {Roman Goldenberg and Ron Kimmel and Ehud Rivlin and Michael Rudzsky},
year = {2005},
journal = {Pattern Recognition},
volume = {38},
number = {8},
pages = {1033-1043},
keywords = {Visual motion; Segmentation and tracking; Object recognition; Activity recognition; Non-rigid motion; Active contours;Periodicity analysis; Motion-based classification; Eigenshapes},
abstract = {We show how periodic motions can be represented by a small number of eigenshapes that capture the whole dynamic mechanism of the motion. Spectral decomposition of a silhouette of a moving object serves as a basis for behavior classification by principle component analysis. The boundary contour of the walking dog, for example, is first computed efficiently and accurately. After normalization, the implicit representation of a sequence of silhouette contours given by their corresponding binary images, is used for generating eigenshapes for the given motion. Singular value decomposition produces these eigenshapes that are then used to analyze the sequence. We show examples of object as well as behavior classification based on the eigendecomposition of the binary silhouette sequence.}
}