Discriminative human action recognition using pairwise CSP classifiers

We present a discriminative approach to human action recognition. At the heart of our approach is the use of common spatial patterns (CSP), a spatial filter technique that transforms temporal feature data by using differences in variance between two classes. Such a transformation focusses on differe...

Full description

Saved in:
Bibliographic Details
Published in2008 IEEE International Conference on Automatic Face and Gesture Recognition pp. 1 - 6
Main Authors Poppe, R., Poel, M.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.09.2008
Subjects
Computer science
Heart
Histograms
Humans
Image recognition
Image representation
Shape
Spatial filters
Video sequences
Video surveillance
Online AccessGet full text

Cover

More Information
Summary:We present a discriminative approach to human action recognition. At the heart of our approach is the use of common spatial patterns (CSP), a spatial filter technique that transforms temporal feature data by using differences in variance between two classes. Such a transformation focusses on differences between classes, rather than on modelling each class individually. As a results, to distinguish between two classes, we can use simple distance metrics in the low-dimensional transformed space. The most likely class is found by pairwise evaluation of all discriminant functions. Our image representations are silhouette boundary gradients, spatially binned into cells. We achieve scores of approximately 96% on a standard action dataset, and show that reasonable results can be obtained when training on only a single subject. Future work is aimed at combining our approach with automatic human detection.
ISBN:1424421535
9781424421534
DOI:10.1109/AFGR.2008.4813447