Body region labelling for action recognition

Dickinson, Patrick and Hunter, Andrew (2006) Body region labelling for action recognition. In: IASTED International Conference on Vizualization, Imaging and Image Processing, 28-30 August 2006, Palma de Mallorca, Spain.

Full text not available from this repository.

Official URL: http://www.actapress.com/Abstract.aspx?paperId=280...

Abstract

We present a novel method for automatically labelling the head, torso, and legs of a human body tracked through a video sequence. An appearance-based body model is constructed by dividing the initial silhouette into a series of spatial slices, and building a colour distribution histogram for each. In subsequent frames a labelling hypothesis is constructed for each new silhouette by matching against these distributions, and used to identify each body region under a range of poses. We use the body model to extract feature points, which we use as the basis for an action recognition scheme. Actions are represented by a vector of the head and torso positions, sampled over the duration of an action. Manually labelled sequences provide a training set comprising sitting, bending, squatting, and lying actions, viewed from various angles. We use nearest-neighbour matching to identify actions presented in test sequences. Our results show that our method is effective, achieving a high recognition rate.

Item Type:Conference or Workshop Item (Paper)
Additional Information:We present a novel method for automatically labelling the head, torso, and legs of a human body tracked through a video sequence. An appearance-based body model is constructed by dividing the initial silhouette into a series of spatial slices, and building a colour distribution histogram for each. In subsequent frames a labelling hypothesis is constructed for each new silhouette by matching against these distributions, and used to identify each body region under a range of poses. We use the body model to extract feature points, which we use as the basis for an action recognition scheme. Actions are represented by a vector of the head and torso positions, sampled over the duration of an action. Manually labelled sequences provide a training set comprising sitting, bending, squatting, and lying actions, viewed from various angles. We use nearest-neighbour matching to identify actions presented in test sequences. Our results show that our method is effective, achieving a high recognition rate.
Keywords:Action recognition, Body region labelling
Subjects:G Mathematical and Computer Sciences > G400 Computer Science
Divisions:College of Science > School of Computer Science
ID Code:105
Deposited By: Patrick Dickinson
Deposited On:01 Sep 2006
Last Modified:18 Jul 2011 16:11

Repository Staff Only: item control page