Object localization method, involves calculating probability values for limitation boxes in input image based on probability that object is arranged in these locations, and selecting box with highest probability as location of object

摘要

The method involves constructing a probability theoretical reinforcing tree, and receiving an input image that contains an object to be localized. Limitation boxes are identified in the image based on a conditional later probability. Probability values are calculated for the boxes based on the probability that the object is arranged in these locations. The box with the highest probability is selected as the location of the object. An independent claim is also included for a method for classifying images of objects into different image categories.