Kinematic jump processes for monocular 3D human tracking

A major difficulty for 3D (three-dimensional) human body tracking from monocular image sequences is the near nonobservability of kinematic degrees of freedom that generate motion in depth. For known link (body segment) lengths, the strict nonobservabilities reduce to twofold 'forwards/backwards flipping' ambiguities for each link. These imply 2/sup # links/ formal inverse kinematics solutions for the full model, and hence linked groups of O(2/sup # links/) local minima in the model-image matching cost function. Choosing the wrong minimum leads to rapid mistracking, so for reliable tracking, rapid methods of investigating alternative minima within a group are needed. Previous approaches to this have used generic search methods that do not exploit the specific problem structure. Here, we complement these by using simple kinematic reasoning to enumerate the tree of possible forwards/backwards flips, thus greatly speeding the search within each linked group of minima. Our methods can be used either deterministically, or within stochastic 'jump-diffusion' style search processes. We give experimental results on some challenging monocular human tracking sequences, showing how the new kinematic-flipping based sampling method improves and complements existing ones.

[1]  Alan H. Barr,et al.  Global and local deformations of solid primitives , 1984, SIGGRAPH.

[2]  Hsi-Jian Lee,et al.  Determination of 3D human body postures from a single view , 1985, Comput. Vis. Graph. Image Process..

[3]  R. Fletcher.  Practical Methods of Optimization , 1988 .

[4]  Claude Samson,et al.  Robot Control: The Task Function Approach , 1991 .

[5]  Robot Control , 1992 .

[6]  Larry S. Davis,et al.  3-D model-based tracking of humans in action: a multi-view approach , 1996, Proceedings CVPR IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[7]  David J. Fleet,et al.  Stochastic Tracking of 3D Human Figures Using 2D Image Motion , 2000, ECCV.

[8]  Camillo J. Taylor,et al.  Reconstruction of Articulated Objects from Point Correspondences in a Single Uncalibrated Image , 2000, Comput. Vis. Image Underst..

[9]  Norman I. Badler,et al.  Real-Time Inverse Kinematics Techniques for Anthropomorphic Limbs , 2000, Graph. Model..

[10]  Andrew Blake,et al.  Articulated body motion capture by annealed particle filtering , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[11]  Michael Isard,et al.  Partitioned Sampling, Articulated Objects, and Interface-Quality Hand Tracking , 2000, ECCV.

[12]  Cristian Sminchisescu,et al.  Covariance scaled sampling for monocular 3D body tracking , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[13]  Ian D. Reid,et al.  Automatic partitioning of high dimensional search spaces associated with articulated body motion capture , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[14]  David J. Fleet,et al.  People tracking using hybrid Monte Carlo filtering , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.

[15]  Cristian Sminchisescu,et al.  Hyperdynamics Importance Sampling , 2002, ECCV.

[16]  Cristian Sminchisescu,et al.  Building Roadmaps of Local Minima of Visual Models , 2002, ECCV.

[17]  Michael J. Black,et al.  Implicit Probabilistic Models of Human Motion for Synthesis and Tracking , 2002, ECCV.

[18]  Cristian Sminchisescu.  Consistency and coupling in human model likelihoods , 2002, Proceedings of Fifth IEEE International Conference on Automatic Face Gesture Recognition.

[19]  Michael Isard,et al.  CONDENSATION—Conditional Density Propagation for Visual Tracking , 1998, International Journal of Computer Vision.