A system (Fig. 7) and a method (Fig. 2) are disclosed for an adaptive discriminative generative model (236) with a probabilistic interpretation. As applied to visual tracking, the discriminative generative model (236) separates the target object from the background more accurately and efficiently than conventional methods. A computationally efficient algorithm constantly updates (248) the discriminative model over time. The discriminative generative model (236) adapts to accommodate dynamic appearance variations of the target and background. According to an alternate embodiment, visual tracking is formulated by defining an object class and one or more background classes. The most discriminant features available in the images are then used to select a portion of each image as belonging to the object class. Fisher’s linear discriminant (1036) method is used to project high-dimensional image data onto a lower-dimensional space, e.g., a line, and perform classification in the lower-dimensional space. The projection function is incrementally updated (1056).
展开▼
