Abstract
Visual tracking, in essence, deals with non-stationary data streams that change over time. While most existing algorithms are able to track ob jects well in controlled environments, they usually fail if there is a significant change in ob ject appearance or surrounding illumina- tion. The reason being that these visual tracking algorithms operate on the premise that the models of the ob jects being tracked are invariant to internal appearance change or external variation such as lighting or viewpoint. Consequently most tracking algorithms do not update the models once they are built or learned at the outset. In this paper, we present an adaptive probabilistic tracking algorithm that updates the models using an incremental update of eigenbasis. To track ob jects in two views, we use an efiective probabilistic method for sampling afine motion parameters with priors and predicting its location with a maxi- mum a posteriori estimate. Borne out by experiments, we demonstrate the proposed method is able to track ob jects well under large lighting, pose and scale variation with close to real-time performance.