Model Driven Segmentation of Articulating Humans in Laplacian Eigenspace

• Published in: IEEE transactions on pattern analysis and machine intelligence Vol. 30; no. 10; pp. 1771 - 1785
• Main Authors: Sundaresan, A., Chellappa, R.
• Format: Journal Article
• Language: English
• Published: Los Alamitos, CA IEEE 01.10.2008; IEEE Computer Society; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Application software; Applied sciences; Articulated; Artificial Intelligence; Biological system modeling; Cameras; Computer science; control theory; systems; Computer Simulation; Exact sciences and technology; Graph-theoretic methods; Graphs; Human; Human body; Humans; Image Enhancement - methods; Image Interpretation, Computer-Assisted - methods; Image Processing and Computer Vision; Image segmentation; Imaging, Three-Dimensional - methods; Joints; Joints - anatomy & histology; Laplace equations; Limbs; Models, Anatomic; Motion estimation; partitioning; Pattern Recognition; Pattern Recognition, Automated - methods; Pattern recognition. Digital image processing. Computational geometry; Region growing; Reproducibility of Results; Segmentation; Sensitivity and Specificity; Splines; Streaming media; Studies; Surveillance; Three dimensional; Whole Body Imaging - methods; partitioning; Pattern Recognition; Segmentation; Graph-theoretic methods; Region growing; Image Processing and Computer Vision; Image processing; Threshold detection; Top down method; Modeling; Spline approximation; Multidimensional analysis; Voxel; region growing; Pattern analysis; Body; Computer vision; Head; Probabilistic approach; Chain length; graph-theoretic methods; Model driven architecture; Graph theory; Pattern recognition; Object recognition; Laplacian; Bottom up method; Tridimensional image; Graph method; Graphic method; Artificial intelligence
• DOI: 10.1109/TPAMI.2007.70823
• ISSN: 0162-8828

We propose a general approach using Laplacian Eigenmaps and a graphical model of the human body to segment 3D voxel data of humans into different articulated chains. In the bottom-up stage, the voxels are transformed into a high-dimensional (6D or less) Laplacian Eigenspace (LE) of the voxel neighborhood graph. We show that LE is effective at mapping voxels on long articulated chains to nodes on smooth 1D curves that can be easily discriminated, and prove these properties using representative graphs. We fit 1D splines to voxels belonging to different articulated chains such as the limbs, head and trunk, and determine the boundary between splines using the spline fitting error. A top-down probabilistic approach is then used to register the segmented chains, utilizing their mutual connectivity and individual properties. Our approach enables us to deal with complex poses such as those where the limbs form loops. We use the segmentation results to automatically estimate the human body models. While we use human subjects in our experiments, the method is fairly general and can be applied to voxel-based segmentation of any articulated object composed of long chains. We present results on real and synthetic data that illustrate the usefulness of this approach.