Previous knowledge, level amounts of representations and visual grouping

摘要

A method for detecting an object in an image comprises the steps of: determining a the form of model training examples of aligned and implement the form of model as a previously known within a level amounts of frame for the determination of the object in the image, wherein the level amounts frame a projection of the object in the image and a confidence measure along of parts of the projection are determined, wherein the level amounts of frame is limited by the previously known.