Segmentation-based Object Categorization - OBJ CUT

OBJ CUT

OBJ CUT is an efficient method that automatically segments an object. The OBJ CUT method is a generic method, and therefore it is applicable to any object category model. Given an image D containing an instance of a known object category, e.g. cows, the OBJ CUT algorithm computes a segmentation of the object, that is, it infers a set of labels m.

Let m be a set of binary labels, and let be a shape parameter( is a shape prior on the labels from a layered pictorial structure (LPS) model). An energy function is defined as follows.

(1)

The term is called a unary term, and the term is called a pairwise term. A unary term consists of the likelihood based on color, and the unary potential based on the distance from . A pairwise term consists of a prior and a contrast term .

The best labeling minimizes, where is the weight of the parameter .

(2)

Read more about this topic:  Segmentation-based Object Categorization

Famous quotes containing the word cut:

    The eating of a MacDonald’s meal is like the reading of Reader’s Digest—small, easily digested, carefully processed, carefully cut down, abridged. Reader’s Digest gives us knowledge that is easily compartmentalized, simplified, ideologically sound.
    Clive Bloom, British educator. “MacDonald’s Man Meets Reader’s Digest,” Readings in Popular Culture: Trivial Pursuits?, St. Martin’s Press (1990)