No abstract

person tracking, activity recognition, pose estimation, stereo camera, eigenpose Plan-view projection of real-time depth imagery can improve the statistics of its intrinsic 3D data, and allows for cleaner separation of occluding and closely-interacting people. We build a probabilistic, realtime multi-person tracking system upon a plan-view image substrate that well preserves both shape and size information of foreground objects. The tracking's robustness derives in part from its "plan-view template" person models, which capture detailed properties of people's body configurations. We demonstrate that these same person models -obtained with a single compact stereo camera unit -may also be used for fast recognition of body pose and activity. Principal components analysis is used to extract plan-view "eigenposes", onto which person models, extracted during tracking, are projected to produce a compact representation of human body configuration. We then formulate pose recognition as a classification problem, and use support vector machines (SVMs) to quickly distinguish between, for example, different directions people are facing, and different body poses such as standing, sitting, bending over, crouching, and reaching. The SVM outputs are transfo rmed to probabilities and integrated across time in a probabilistic framework for real-time activity recognition.

Early stage cerebral aneurysms are characterized by the disruption of the internal elastic lamina. The cause of this breakdown is still not understood, but it has been conjectured to be due to fatigue failure and/or by a breakdown in homeostatic mechanisms in the wall arising from some aspect of the local hemodynamics and wall tension. We propose to model this disruption using a structural damage model. It is built on a previously introduced nonlinear, inelastic multi-mechanism model for cerebral arteries (2005, "An Inelastic Multi-Mechanism Constitutive Equation for Cerebral Arterial Tissue," Biomech. Model. Mechanobiol., 4(4), pp. 235-248), as well as a recent generalization to include the wall anisotropy (2009, "A Structural Multi-Mechanism Constitutive Equation for Cerebral Arterial Tissue," Int. J. Solids Struct., 46(14-15), pp. 2920-2928). The current model includes subfailure damage of the elastin, represented by changes in the tissue mechanical properties and unloaded reference length. A structural model is used to characterize the gradual degradation, failure of elastin, and recruitment of anisotropic collagen fibers. The collagen fibers are arranged in two helically oriented families with dispersion in their orientation. Available inelastic experimental data for cerebral arteries are used to evaluate the constitutive model. It is then implemented in a commercial finite element analysis package and validated using analytical solutions with representative values for cerebral arterial tissue.

No abstract

a b s t r a c tA structural multi-mechanism constitutive equation is developed to describe the nonlinear, anisotropic, inelastic mechanical behavior of cerebral arterial tissue. Elastin and collagen fibers are treated as separate components (mechanisms) of the artery. Elastin is responsible for load bearing at low strain levels while the collagen mechanism is recruited for load bearing at higher strain levels. This work builds on an earlier model in which both the elastin and collagen mechanisms are represented by isotropic response functions [Wulandana, R., Robertson, A.M., 2005. An inelastic multi-mechanism constitutive equation for cerebral arterial tissue. Biomech. Model. Mechan. 4 (4), 235-248]. Here, the anisotropic material response of the wall is introduced through the collagen mechanism which is composed of helically distributed families of fibers. The orientation of these families is described using either a finite number of fiber orientations or a fiber distribution function. The fiber orientation or dispersion function can be prescribed directly from arterial histology data, or, taking a phenomenological approach, based on data fitting from bi-axial measurements. The activation of the collagen mechanism is specified using a new fiber strain based activation criterion. The multi-mechanism constitutive equation is applied to the simple case of cylindrical inflation and material constants are determined based on available inelastic experimental data for cerebral arteries. While the proposed model captures all features of this inelastic data, there is a pressing need for further experiments to refine the model.