Modelling Human Shape with Articulated Shape Mixtures
This paper describes a statistical framework for recognising 2D shapes with articulated components. The shapes are represented using both geometrical and a symbolic primitives, that are encapsulated in a two layer hierarchical architecture. Each primitive is modelled so as to allow a degree of articulated freedom using a polar point distribution model that captures how the primitive movement varies over a training set. Each segment is assigned a symbolic label to distinguish its identity, and the overall shape is represented by a configuration of labels. We demonstrate how both the point-distribution model and the symbolic labels can be combined to perform recognition using a probabilistic hierarchical algorithm. This involves recovering the parameters of the point distribution model that minimise an alignment error, and recovering symbol configurations that minimise a structural error. We apply the recognition method to human moving skeleton.
KeywordsRecognition Rate Modelling Human Training Pattern Posteriori Probability Alignment Error
- 1.Cootes, T., Taylor, C.: Combining point distribution models with shape models based on finite element analysis. IVC 13(5), 403–409 (1995)Google Scholar
- 2.Duta, N., Jain, A., Dubuisson, P.: Learning 2d shape models. International Conference on Computer Vision and pattern Recognition 2, 8–14 (1999)Google Scholar
- 4.Gonzales, J., Varona, J., Roca, F., Villanueva, J.: aspace: Action space for recognition and synthesis of human actions. In: 2nd IWAMDO, Spain, pp. 189–200 (2002)Google Scholar
- 10.Hancock, E.R., Kittler, J.: Edge-labelling using dictionary-based relaxation. IEEE Transaction on PAMI 12(2), 165–181 (1990)Google Scholar