Stochastic Systems Group
Home Research Group Members Programs  
Demos Calendar Publications Mission Statement Alumni

SSG Seminar Abstract


Combining Approximate Dynamical Models with Approximate Inference in Loopy Graphs


Leonid Taycher - CSAIL, MIT


Stochastic tracking of structured models in monolithic state spaces often requires modeling complex distributions that are difficult to represent with either parametric or sample-based approaches. However, multiple simpler models are often available, each of which captures some aspect of the complex behavior. For example, human body parts may be robustly tracked individually, but the resulting pose combinations may not satisfy articulation constraints. Conversely, the results produced by full-body trackers satisfy such constraints, but such trackers are usually fragile due to the presence of clutter.

We propose a method for combining such redundant models to improve individual state estimates. The constituent models are combined in a manner similar to a Product of HMMs model. Hidden variables are introduced to represent system appearance. While the resulting model contains loops, making the inference hard in general, we present an approximate non-loopy filtering algorithm based on sequential application of Belief Propagation to acyclic subgraphs of the model.



Problems with this site should be emailed to jonesb@mit.edu