Human Motion Segmentation Using Active Shape Models
Human motion analysis from images is meticulously related to the development of computational techniques capable of automatically identifying, tracking and analyzing relevant structures of the body. This work explores the identification of such structures in images, which is the first step of any computational system designed to analyze human motion. A widely used database (CASIA Gait Database) was used to build a Point Distribution Model (PDM) of the structure of the human body. The training dataset was composed of 14 subjects walking in four directions, and each shape was represented by a set of 113 labelled landmark points. These points were composed of 100 contour points automatically extracted from the silhouette combined with an additional 13 anatomical points from elbows, knees and feet manually annotated. The PDM was later used in the construction of an Active Shape Model, which combines the shape model with gray level profiles, in order to segment the modelled human body in new images. The experiments with this segmentation technique revealed very encouraging results as it was able to gather the necessary data of subjects walking in different directions using just one segmentation model.
KeywordsContour Point Landmark Point Segmentation Model Stick Point Stick Figure
The first author would like to thank the support of the Ph.D. grant with references SFRH/BD/28817/2006 from Fundação para a Ciência e Tecnologia (FCT), in Portugal. This work was partially developed under the scope of the project with reference PTDC/BBB-BMD/3088/2012 financially supported by FCT. The research described in this chapter use CASIA Gait Database collected by Institute of Automation, Chinese Academy of Sciences.
- 1.Ogawara K, Li X, Ikeuchi K. Marker-less Human Motion Estimation using Articulated Deformable Model. IEEE International Conference on Robotics and Automation, 2007. p. 46–51.Google Scholar
- 4.Das Choudhury S, Tjahjadi T. Gait recognition based on shape and motion analysis of silhouette contours. Computer Vision and Image Understanding, 2013. 117(12): 1770–85.Google Scholar
- 5.Tran C, Trivedi MM. Human body modelling and tracking using volumetric representation: Selected recent studies and possibilities for extensions. Second ACM/IEEE International Conference on Distributed Smart Cameras, 2008. p. 1–9.Google Scholar
- 7.Cootes TF, Taylor CJ, Cooper DH, Graham J. Training Models of Shape from Sets of Examples. BMVC92. Springer London, 1992. p. 9–18.Google Scholar
- 8.Jang C, Jung K. Human pose estimation using Active Shape Models. Proceedings of World Academy of Science: Engineering & Technology 46 (2008).Google Scholar
- 9.Pourjam E, Ide I, Deguchi D, Murase H. Segmentation of Human Instances Using Grab-cut and Active Shape Model Feedback. Proceedings of IAPR Conference on Machine Vision Applications (MVA) 2013. p. 77–80.Google Scholar
- 11.Nixon MS, Tan T, Chellappa R. Human Identification Based on Gait. Vol 4, Springer; 2010.Google Scholar
- 12.Yu S, Tan D, Tan T. A Framework for Evaluating the Effect of View Angle, Clothing and Carrying Condition on Gait Recognition. 18th International Conference on Pattern Recognition, 2006. p. 441–444.Google Scholar
- 13.Cootes TF, Taylor CJ, Lanitis A. Multi-resolution search with active shape models. Proceedings of the 12th IAPR International Conference on Pattern Recognition, 1994. Vol.1 p. 610–612.Google Scholar
- 14.Hamarneh G. Active Shape Model Software [Internet]. 1999. Available from: http://www.cs.sfu.ca/~hamarned/software/code/asm.zip Last accessed June 2012.