Object classification through semantic mapping

Number: 9740964
Published: 2017-08-22
Filed: 2015-03-31
Assignee: Disney Enterprises, Inc.
Inventors: Hwang; Sung Ju et al.
CPC: G06F18/2431; G06V10/764; G06F18/2451; G06F18/2411
Verdict: Set aside object classification via semantic mapping - generic content ML
Source: Google Patents · FreePatentsOnline

Abstract

There are provided systems and methods for performing object classification through semantic mapping. Such an object classification system includes a system processor, a system memory, and an object categorizing unit stored in the system memory. The system processor is configured to execute the object categorizingunit to receive image data corresponding to an object, and to transform the image data into a directed quantity expressed at least in part in terms of semantic parameters. The system processor is further configured to determine a projection of the directed quantity onto an object representation map including multiple object categories, and to associate the object with a category from among the multiple object categories based on the projection.

Background

BRIEF DESCRIPTION OF THE DRAWINGS(1) FIG. 1 shows a diagram of an exemplary object classification system, according to one implementation;(2) FIG. 2 shows another exemplary implementation of an object classification system;(3) FIG. 3 shows an exemplary system and a computer-readable non-transitory medium including instructions for performing object classification through semantic mapping;(4) FIG. 4 is a flowchart presentingan exemplary method for use by a system to perform object classification through semanticmapping; and(5) FIG. 5 shows an exemplary diagram illustrating association of an object with a category based on a supercategory to which the category belongs and one or more attributes of the object.DETAILED DESCRIPTION(6) The following description contains specific information pertaining to implementations in the present disclosure. One skilled in the artwill recognize that the present disclosure may be implemented in a manner different from that specifically discussed herein. The drawings in the present application and their accompanying detailed description are directed to merely exemplary implementations. Unless noted otherwise, like or corresponding elements among the figures may be indicated by like orcorresponding reference numerals. Moreover, the drawings and illustrations in the presentapplication are generally not to scale, and are not intended to correspond to actual relative dimensions.(7) As noted above, object classification is a challenging problem

Claims

1. An object classification system comprising: a system processor; a system memory; and an object categorizing unit stored in the system memory; wherein the system processor is configured to execute the object categorizing unit to: receive image data corresponding to an object; transform the image data into a directed quantity expressed at least in part in terms of semantic parameters; determine a projection of the directed quantity onto an object representation map including a plurality of object categories; associate the object with a category from among the plurality of object categories based on the projection; and issue at least one query to a user for adapting the object representation map based on a ranked list of semantic relationships of the plurality of object categories. 2. Theobject classification system of claim wherein the system processor is configured to execute the object categorizing unit to associate the object with the category based on a proximity of the projection to the category on the object representation map. 9. A method for use by a object classification system including a system processor and an object categorizing unit stored in a system memory, the method comprising: receiving image data corresponding to an object; transforming the image data into a directed quantity expressed at least in part in terms of semantic parameters; determining a projection of the directed quantity onto an object representation map including a plurality of object categories; associating the object with a category from among the plurality of object categories based on the projection; and issue at least one query to a user for adapting the object representation map based on a ranked list of semantic relationships of the plurality of object categories. 17. A computer-readable non-transitory medium having stored thereon instructions, which whenexecuted by a system processor, instantiate a method comprising: receiving image data corresponding to an object; transforming the image data into a directed quantity expressed atleast in part in terms of semantic parameters; determining a projection of the directed quantity onto an object representation map including a plurality of object categories; associating the object with a category from among the plurality of object categories based on the projection; and issuing at least one query to a user for adapting the object representation map based on a ranked list of semantic relationships of the plurality of object categories.