No abstract

No abstract

An approach is presented to match imaged trajectories of anatomical landmarks (e.g. hands, shoulders and feet) using semantic correspondences between human bodies. These correspondences are used to provide geometric constraints for matching actions observed from different viewpoints and performed at different rates by actors of differing anthropometric proportions. The fact that the human body has approximate anthropometric proportion allows innovative use of the machinery of epipolar geometry to provide constraints for analyzing actions performed by people of different sizes, while ensuring that changes in viewpoint do not affect matching. In addition, for linear time warps, a novel measure, constructed only from image measurements of the locations of anatomical landmarks across time, is proposed to ensure that similar actions performed at different rates are accurately matched as well. An additional feature of this new measure is that two actions from cameras moving at constant (and possibly different) velocities can also be matched. Finally, we describe how dynamic time warp-A. Gritai ( ) Cernium Corporation, ing can be used in conjunction with the proposed measure to match actions in the presence of nonlinear time warps. We demonstrate the versatility of our algorithm in a number of challenging sequences and applications, and report quantitative evaluation of the matching approach presented.

We propose a novel approach for tracking of human joints based on anthropometric constraints. A human is modeled as a pictorial structure consisting of body landmarks (joints) and corresponding links between them. Anthropometric constraints relate the landmarks of two persons if they are in the same posture. Given a test video, where an actor performs the same action as in a model video, and joint locations in the model video, anthropometric constraints are used to determine the epipolar lines, where the potential joint locations are searched in the test video. The edge templates around joints and related links are used to locate joints in the test video. The performance of this method is demonstrated on several different human actions.

No abstract